Investigating and Litigating Wildfire Origin, Cause, and Loss Cases

Front matter

Copyright and contents

Contents

1. Introduction: Every Fire Has an Owner
2. Chapter 1: Four Cases in One
3. Chapter 2: How a Fire Starts
4. Chapter 3: Reading the Ground
5. Chapter 4: Cause Determination and the Rulebooks
6. Chapter 5: Weather, Fuel, and the Record
7. Chapter 6: Evidence and Spoliation
8. Chapter 7: Getting Your Expert Admitted
9. Chapter 8: Utility Liability
10. Chapter 9: Who Actually Pays
11. Chapter 10: Criminal Exposure
12. Chapter 11: The Marshall Fire
13. Chapter 12: Eaton and Palisades
14. Chapter 13: The Destroyed Home
15. Chapter 14: When the View Burns and the House Does Not
16. Chapter 15: Smoke, Ash, and Char
17. Chapter 16: The Downstream Years
18. Chapter 17: The Insurance Gap
19. Chapter 18: The Human Cost
20. Chapter 19: Trying the Case
21. Appendix A: Glossary
22. Appendix B: Sources by Chapter
23. Appendix C: Verification Punch List
24. Appendix D: About the Author

Introduction

Every Fire Has an Owner

Every fire in this book started with somebody's equipment or somebody's activity. A power line, a burn pile, a truck exhaust, a mower blade striking a rock. Heat escaped the equipment that was supposed to contain it and reached dry fuel.

Four questions then decide the case. Where did the fire start. What started it. Who is responsible. What was lost. The first three have a literature, a standard, and a trained profession behind them. The fourth has none of that, and it is where most cases leave money on the table. About a third of this book sits there.

Wildfire harm travels far past the burn. The house that did not burn lost its view, its air, and its street. The hillside above town floods for five years. The vineyard downwind sells its grapes at a discount or dumps them. The owner insured to a number an agent produced in fifteen minutes learns what that number left out. None of it shows up in the origin and cause report. All of it is compensable if somebody proves it.

In a utility case, the utility controls the physical evidence and the utility is the defendant. Its crew reaches the pole before your investigator does, cuts the conductor, bags it, and hauls it to a warehouse. Everything you will ever learn about that hardware passes through them first. Send the preservation letter the day you are retained, and negotiate a joint examination protocol before anything moves.

A destroyed home is the easy case. A surviving home with a burned ridgeline, lead in the carpet, and eleven empty lots on the block is the hard one, and it is worth more than most lawyers claim for it.

The law keeps moving. California rebuilt its cost allocation system in 2019 and expanded it in 2025. Seven states passed utility favorable liability statutes in the 2025 session alone. Where a figure here is reported rather than verified, the text says so and names who reported it. Nothing here substitutes for the primary document.

Chapter 1

Four Cases in One

A wildfire case is four cases stacked on each other, tried at once, by people with different training. Where the fire started is a physical evidence case. What started it is an engineering case. Who is responsible is a regulatory and tort case. What was lost is a valuation case. Lose any one and the other three do not matter. The fourth gets underworked, and that habit costs clients more than every evidentiary ruling in the case combined.

The four questions

Where it started. A search problem before it is an opinion problem. The investigator works from the area of most intense burning back toward the area of least, normally to the heel, the slowest burning edge. If the first arriving engine company did not protect and flag the scene, nothing downstream is reliable.

What started it. A competent ignition source in contact with a receptive first fuel: how big a particle, how hot, how dry the fuel. The defense argument in almost every utility case is that the arc could not have thrown a particle large enough and hot enough to ignite that fuel at that moisture. That is testable, which is what the courts now demand.

Who is responsible. In California, an investor owned utility can be liable in inverse condemnation with no showing of negligence. In Oregon the same facts require proof of negligence, and produced a jury finding of willful conduct with punitive damages. In Texas a utility that complied with an approved wildfire mitigation plan and was not negligent has a statutory safe harbor. The same conductor failure produces radically different exposure depending on which side of a state line the pole stands on.

What was lost. Destroyed homes, damaged homes, and untouched homes with a destroyed setting. Trees, contents, business income, agricultural crops. Evacuation costs, loss of use, annoyance and discomfort. And the multi year debris flow hazard on a burn scar. Each has its own measure of damages, and several have their own statutes.

Who is in the room

Wildfire litigation is mass litigation with an unusual mix of claimants.

Pacific Gas and Electric Company settled with insurance subrogation plaintiffs for eleven billion dollars on September 13, 2019, more than it paid the eighteen public entities and comparable to the trust for individual victims. Hawaii went the other way. The Hawaii Supreme Court held in In re: The Petition for the Coordination of Maui Fire Cases, No. SCRQ-24-0000602, that the statutory lien under Hawaii Revised Statutes section 663-10(a) is "the exclusive remedy for a property and casualty insurer to recover claims paid for damages caused by a third-party tortfeasor in the context of a tort settlement." Carriers who had paid roughly $3.03 billion were projected to recover about $216 million, near ten cents on the dollar.

Why the fourth question is where the money is left

One year after the January 2025 Los Angeles fires, homes inside the burn footprint that did not burn and had no smoke claim lost 9.6 percent of aggregate value in Pacific Palisades and 15.8 percent in Altadena. No policy covers that, no adjuster asks, and in most cases nobody pleads it. On top of it sits the insurance gap: 69 percent of surveyed households reported being underinsured, 7 percent adequate coverage. Chapters fourteen and seventeen work both through.

There is a barely hidden nationwide crisis of underinsurance.

Kenneth S. Klein, Louis and Hermione Brown Professor of Law, California Western School of Law, "Truth and Consequences: What Catastrophe Teaches Us About Homeownership and Underinsurance," Lewis & Clark Law Review, Vol. 30.1, 2026

The order of operations that keeps a case alive

The evidence that decides causation belongs to the defendant, sits on its property, and is about to be hauled away by crews under real pressure to restore power. Send the preservation letter before the complaint.

1. Preservation letter naming physical components and electronic records, within days.
2. Joint evidence examination protocol negotiated before any component is cleaned, cut, or shipped.
3. Weather and fuel record pulled from public archives before stations are decommissioned or data rolls off.
4. Origin and cause expert retained, and separately, an electrical engineer if a utility is in the case.
5. Damages experts retained early, not after liability is resolved. Diminution in value opinions need pre fire comparable sales, and the comparable set degrades every month after the fire.

Chapter 2

How a Fire Starts

Ignition is a heat transfer problem with measurable thresholds. A lawyer who knows them can cross examine an ignition opinion. A lawyer who does not will take "the arc caused the fire" or "the arc could not have caused the fire" as a conclusion rather than a hypothesis.

The fire triangle, and why the third leg is the one in dispute

Combustion requires fuel, oxygen, and heat. Oxygen is never in issue in a wildfire case and fuel rarely is. The fight is over heat: whether the asserted source delivered enough energy, long enough, into fuel dry enough to sustain flaming combustion.

Hot metal particles have a measured ignition boundary

Urban, Zak and Fernandez-Pello dropped aluminum particles from 2 to 8 millimeters in diameter into natural fuel beds and mapped an ignition boundary in particle size and temperature. The temperature required rises as the particle gets smaller, because a small particle carries less stored energy and sheds it faster: a 2 millimeter particle must be substantially hotter than an 8 millimeter particle to ignite the same fuel. Fuel chemistry moves the boundary. Pure alpha cellulose ignited at lower particle temperatures than natural biomass, because lignin and hemicellulose in grass blends and pine needle beds raise the threshold, and powdered fuels ignite lower than coarser beds of the same material.

The authors also found that fires started by clashing conductors grow faster than other fires, because the same high wind that swung the conductors together drives the spread of what they ignite.

Six ways a power line starts a fire

The Texas A&M Engineering Experiment Station's wildfire mitigation program publishes the clearest public taxonomy. Each mechanism leaves a different signature and implies different records.

Downed conductor. An energized conductor breaks and falls. In roughly 30 percent of cases the protective devices do not operate, because fault current through dry soil, vegetation, or asphalt is below the relay pickup setting, so the conductor stays energized and arcs at high energy for tens of minutes, often until a customer calls in an outage. Look for the conductor in fuels, arc erosion and beading at the contact point, locally fused soil, a cone of ejected metal spatter, and pole top or dead end failure upstream.

Conductor to vegetation contact. A limb bridges conductor to ground or conductor to conductor, ignites, arcs, and eventually burns the conductor in two; the same source documents intermittent contact over 24 hours before the conductor severed. Look for an arc charred branch with a vitrified contact point, arc erosion pits on the conductor at a matching elevation, and the tree species and height measured against conductor sag, then get the vegetation management records for that span.

Conductor slap. Conductors swing into each other, arc, and eject hot metal particles into ground fuels below. In fault induced conductor slap, a fault at one location swings conductors together up to a mile away, creating a secondary fault, and lines that experience it tend to repeat unless corrected. Look for paired arc marks at matching points on two phases, a scatter field of solidified metal spheres beneath, and no vegetation contact.

Splice and connector failure. Equipment arcs at low level for a pre failure period that intensifies over time, then either ignites combustibles directly or burns the conductor in two. Compression splices with internal voids, hot line clamps, stirrups, and stress cone connectors are all named in the National Wildfire Coordinating Group evidence list. Look for annealing and loss of temper in the conductor near the splice, and oxidation coloration.

Transformer failure. Case failure, bushing flashover, tank rupture, and ejection of burning mineral oil onto ground fuels. Look for oil saturation of soil and ash, a ruptured tank seam, a blown pressure relief, and arc tracking on bushings. Teardown is destructive and cannot be undone, so put it in the protocol.

Wildlife contact. A carcass bridging phase to ground, with characteristic entry and exit burns. The National Wildfire Coordinating Group lists remains showing electrical charring as powerline evidence.

Re-energization is a second ignition event

Automatic reclosing, and manual restoration after an outage, reapply fault current to a conductor already down or in vegetation, producing a second arcing event frequently larger than the first. The record set is relay and supervisory control and data acquisition data, recloser logs, protection settings, and whether reclosing was disabled during elevated fire conditions. Repeated identical faults on the same span give you the notice element of a negligence case out of the defendant's own monitoring system.

Vehicles, equipment, and the roadside pattern

The National Wildfire Coordinating Group treats equipment and vehicle use as a separate cause class, with distinctive evidence.

• Catalytic converters and exhaust. Friable ceramic honeycomb substrate fragments ejected from a deteriorating converter, found in the ignition area, plus carbon and exhaust deposits. Converter shell temperatures under load ignite fine dead fuels on contact when a vehicle is parked or driven over cured grass.
• Spark arrestor failure. Check the arrestor screen on chainsaws, off highway vehicles, chippers, and mowers. A missing or holed screen is both physical evidence and a regulatory violation.
• Rock strike marks. The signature of a mower blade, disc, brush cutter, dozer track, or grader blade striking rock. Look for fresh bright metal transfer on the rock and a matching nick in the implement.
• Dragging chains and metal. Scrape marks on pavement from a safety chain, muffler, or trailer tongue, producing a linear series of small ignitions along a roadway at intervals matching the spark showers.
• Smoldering material on hot exhaust. Packed dry grass on an off highway vehicle exhaust that falls off and keeps smoldering somewhere else.
• Railroad. Carbon material from exhaust systems, brake shoe fragments, hot journal debris, and welding or grinding from maintenance of way. Multiple ignitions strung along a right of way at consistent spacing is the classic railroad signature.

Lightning is either in the case or it is not

Lightning leaves strike marks on trees and objects, splintered wood thrown radially from the strike tree, a needle shower under the crown, a bark explosion spiraling down the trunk along the cambium, fulgurites in sandy soil, and root follow burns. A lightning ignition can smolder in duff for days to weeks before flaring, which is called a holdover or sleeper fire.

The corroborating dataset is the National Lightning Detection Network, operated by Vaisala, which produces time stamped, geolocated cloud to ground strike records with polarity and peak current. No cloud to ground strike inside a defensible radius and time window, holdover included, and no strike tree evidence, and lightning is out. A strike with long continuing current inside the general origin area in the two weeks before discovery puts lightning back in, and it has to be eliminated before declaring a human cause.

The remaining human causes, and their signatures

Chapter 3

Reading the Ground

Origin determination is a search conducted backward. Fire burns hottest and cleanest running with the wind and uphill, and leaves the most unburned material backing into the wind or downslope, so the investigator follows the progression from most intense burning toward least, because the origin normally sits near the heel.

General, specific, and ignition, in that order

"Macro origin area" is not National Wildfire Coordinating Group terminology. Macro and micro modify indicators, not areas. Macro indicators are large object effects: shrubs, trees, char on trunks, crown scorch, the burn pattern visible from the air. Micro indicators are small object effects: grass stems, pebbles, cans, wire, seed heads. A report using "macro origin area" is using colloquial language and should say so.

Protect large, then shrink

The National Wildfire Coordinating Group first responder course teaches one rule that decides cases: protect a larger area than you think you need. An investigator can shrink a protected area but cannot expand one that was too small. First responders flag before the qualified investigator arrives, and the whole origin determination rides on whether that first flagging happened and was documented.

The search pattern is a lane search, left to right and right to left, from more intense burning toward less. Once the general origin area is bounded and protected, the crew grids the specific origin and ignition areas on hands and knees, using metal detectors, rare earth magnets, sifting screens, and thermal imagers.

You cannot read an indicator without the spread context

The eleven indicators, and what each one actually tells you

The National Wildfire Coordinating Group glossary defines a fire pattern indicator as "a physical object that displays changes (fire effects) from exposure to heat, flame, and combustion by-products that can reveal the direction of fire progression at a precise location with accurate analysis," and treats it as one component within a broader fire pattern analysis. There are eleven recognized categories.

1. V and U patterns. A ground level, horizontal burn pattern opening outward in the direction of spread, apex pointing back toward the origin. A narrow V indicates higher wind speed or steeper slope, a broad U lower wind and gentler slope.
2. Grass stem. Unconsumed grass stems and seed heads point in the direction the fire came from, bent back toward the origin as the flame front passes.
3. Protection, also called heat shadowing. An object shields the ground or another object from radiant heat and flame. The sheltered side shows less char, soot, and ash; the exposed side faces the oncoming fire. It is the most used indicator and the one most sensitive to an object being moved after the fire.
4. Sooting. Particulate carbon deposits are heavier on the exposed, fire facing surface.
5. Staining. Vaporized oils and resins driven off by heat condense preferentially on the exposed surface, producing a brown or amber stain. Staining is condensed volatiles, sooting is particulate carbon, and the two are routinely confused.
6. White ash deposition. White ash, the residue of complete combustion, accumulates and adheres more heavily on the exposed surface.
7. Cupping. A cup shaped char pattern on the side of a fuel exposed to the oncoming fire, concavity facing the direction the fire came from. Read on woody stems and small branches.
8. Curling. Heat induced curl of leaves and fine fuels toward the heat source. The National Wildfire Coordinating Group excludes it from first responder training because it is the least reliable of the set.
9. Angle of char. The char line on a vertical fuel rises on the side the fire approached from and slopes downward on the lee side, producing a measurable angle. Deepest char faces the incoming fire.
10. Foliage freeze. Needles and small branches swept by the convective wind of the passing flame front and fixed in that position by rapid desiccation. Simeoni's experimental work found that crown freeze indicated the direction of fire spread, not ambient wind direction, contradicting much common field guidance.
11. Spalling. Chipping, flaking, or fracturing of rock and concrete surfaces from thermal stress, concentrated on the exposed surface.

Logged alongside these is the die-out pattern, the tapering fingers of unburned or lightly burned fuel at the margins where fire ran out of energy or fuel. Die-out fingers point back toward the direction of travel and bound the fire's early, low intensity perimeter.

The wildland V is not the structure V

This trap catches experts who came to wildland work from structure work. The structure fire V is a vertical pattern on a vertical surface, usually a wall, generated by the buoyant plume rising off a fuel package; its apex points down toward the base of the fuel and its angle is governed by heat release rate, ventilation, and fuel package geometry. The wildland V is a horizontal, plan view pattern across the ground, generated by lateral spread; its apex points back toward the origin and its angle is governed by wind speed and slope. Two patterns, two sets of physics, one letter.

The validation problem, stated honestly

Simeoni and colleagues, 2017. An instrumented experimental burn in the New Jersey Pine Barrens in March 2016, documented with video, infrared cameras, and thermocouples. Eight of eleven indicators were observable, and the results were mixed. Directionality varied widely. Microscale indicators tracked local surface conditions more than fire direction, while macroscale indicators did better in high intensity zones. Wind vortex flame wrap and spotting produced directional inconsistencies. Very intense burning destroyed patterns outright, and rain after the fire washed off sooting. The conclusion: read indicators through a general analysis of fire behavior, not in isolation.

Parker and Babrauskas, 2024, "Validation of NWCG Wildfire Directional Indicators in Test Burns in Coastal California," Fire 2024, 7(1), 5. Four experienced investigators independently assessed 404 artifacts across three controlled burn plots, with single point ignition, mild conditions, level terrain, and ground truth from drone video. The average directional error was 103 degrees against a 90 degree random baseline, and six indicators with adequate data ran 81 to 106 degrees. The authors concluded that "such a large error band will not enable the investigator to triangulate back to the correct origin."

The counterweight. The Wildland Fire Investigation Subcommittee published a position paper through the International Association of Wildland Fire arguing that indicators remain useful when read in clusters inside a global fire behavior analysis, citing successful field applications and naming research gaps it acknowledges. It also describes the Parker and Babrauskas paper as having validated investigator reliability. The paper says the opposite. Read both primary sources before relying on either summary, and use the discrepancy on cross.

The defensible position is convergence. An origin determination rests on four legs: fire behavior analysis given documented weather, fuel, and topography; first arrival and witness observations with times; clustered, mutually corroborating indicators; and burn pattern geometry at macro scale. Never a single artifact. Never photographs alone.

Chapter 4

Cause Determination and the Rulebooks

Three documents define competent practice. The National Fire Protection Association's Guide for Fire and Explosion Investigations, known as NFPA 921, sets the method. Its Standard for Professional Qualifications for Fire Investigator, NFPA 1033, sets who is qualified. The National Wildfire Coordinating Group's Guide to Wildland Fire Origin and Cause Determination, publication PMS 412, is the federal wildland analogue to 921.

NFPA 921 is a guide, and courts treat it as the benchmark anyway

The 2024 edition of NFPA 921 issued December 21, 2023 and runs 29 chapters. Chapter 27 is the wildfire chapter, beginning at printed page 921-366, covering wildfire terminology, basic methodology, fire science, fire pattern indicators, safety, determining the origin area, and determining the cause. Chapter 18 covers origin determination generally and Chapter 19 fire cause determination. Its administration chapter says it is a guide rather than a standard, which cuts both ways on cross: deviation is not automatically disqualifying, and an expert who claims 921 compelled a result has overstated it.

The sixteen knowledge areas are a voir dire checklist

NFPA 1033, in the widely cited 2014 edition at section 1.3.7, requires current knowledge above the high school level in sixteen named subjects: fire science, fire chemistry, thermodynamics, thermometry, fire dynamics, explosion dynamics, computer fire modeling, fire investigation, fire analysis, fire investigation methodology, fire investigation technology, hazardous materials, failure analysis and analytical tools, fire protection systems, evidence documentation collection and preservation, and electricity and electrical systems. Section 1.3.8 requires currency through formal education, workshops, seminars, professional publications, and training. Opposing counsel uses that list as a sixteen question voir dire. Prepare your expert to answer each with a specific course, publication, or project, and ask the same sixteen of theirs.

The current published edition is 2022. A 2026 cycle revision in process renumbers the knowledge list to 4.1.7, reorganizes it into four groupings, and adds a prerequisite of at least 32 hours of tested training. Whether the 2026 edition has issued is in the verification punch list.

Get the title of PMS 412 right

The current document is the NWCG Guide to Wildland Fire Origin and Cause Determination, PMS 412, NFES 001874, March 2025 version, announced April 8, 2025, superseding the April 2016 edition. The older title, Wildfire Origin and Cause Determination Handbook, belongs to the legacy National Wildfire Coordinating Group Handbook 1, PMS 412-1, from 1978 and revised in 2005. Do not cite the legacy title for the current document.

Federal and most state wildland investigators are trained to PMS 412 through three courses: FI-110, wildland fire observations and origin scene protection for first responders; FI-210, origin and cause determination; and FI-310, case development. Ask any wildland origin and cause witness which of the three they completed, and when.

The seven steps, and the two kinds of reasoning inside them

Chapter 4 of NFPA 921 relates fire investigation to the scientific method in seven steps: recognize the need, define the problem, collect data, analyze the data, develop a hypothesis, test the hypothesis, select the final hypothesis. Hypothesis development is inductive, moving from observed particulars to a general explanation. Testing is deductive: reason from the hypothesis to what must be true if it holds, then check each consequence against all known facts and, where possible, physical experiment. Chapter 19 mirrors the sequence for cause: data collection, data analysis, developing cause hypotheses, testing for validity, selecting the final hypothesis, and classification.

An expert who cannot say which step was inductive and which was deductive has recited the method rather than performed it. In Indiana Insurance Co. v. General Electric Co., 326 F. Supp. 2d 844 (N.D. Ohio 2004), the opinion was excluded for failure to properly identify physical evidence despite nominal adherence to the steps of NFPA 921.

Bias is a named defect, not a personality trait

Expectation bias is reaching a conclusion before or without adequate data, then interpreting the data to fit it. Confirmation bias is seeking or crediting only the data that confirms the working hypothesis. NFPA 921 prohibits both, and the 2024 edition added express guidance on confirmation bias and on expressions of certainty. The guide also directs the investigator to avoid presumption: no hypothesis is accepted until tested. The questions on cross are when the opinion was first formed, what was known at that moment, and what data arrived afterward that was not revisited.

Undetermined is a conclusion, not a failure

NFPA 921 treats undetermined as a legitimate and required classification. Where two or more ignition scenarios remain equally plausible and the data does not permit eliminating one in favor of another, the cause must be classified undetermined. The Bureau of Land Management states the rule operationally: causes must reach 51 percent probability to be determined, and insufficient evidence results in undetermined classifications rather than speculation.

The California Department of Forestry and Fire Protection, known as CAL FIRE, still lists the cause of the Eaton Fire as undetermined more than eighteen months after the fire. Southern California Edison told securities regulators that a viable explanation is that a de-energized idle transmission facility in the preliminary area of origin was associated with the ignition. Undetermined by the agency and probable in the defendant's own filing are not a contradiction. They are different questions asked to different standards.

Negative corpus was repudiated, and it matters most here

The 2011 edition of NFPA 921 repudiated the negative corpus method: declaring a cause, typically incendiary, purely by eliminating all other causes without affirmative evidence of the asserted ignition source. Pure elimination is the plaintiff's default move against a utility. "Nothing else could have started it, so the power line did" is negative corpus wearing a hard hat, and it needs affirmative evidence: the arc erosion, the metal spatter field, the relay record, the fault distance calculation.

The certainty vocabulary courts accept

Avoid "reasonable degree of scientific certainty." It has no agreed definition, the Department of Justice and the National Commission on Forensic Science have both recommended against it, NFPA 921 cautions against it, and the 2023 amendment to Federal Rule of Evidence 702 points the same way. Experts should not claim absolute or one hundred percent certainty, should give known error rates where those exist, and must keep the opinion inside what a reliable application of the method supports.

Chapter 5

Weather, Fuel, and the Record

The weather on the day is half the causation case and most of the notice case. It decides whether a hot particle that would ordinarily die becomes a fire, and whether a reasonable operator knew it. Almost all of it is publicly archived and free, and almost none is in the fire agency's report.

Humidity drives fuel moisture, and fuel moisture decides ignition

Dead fuels are classified by timelag: 1, 10, 100, and 1000 hour. The timelag is how long a fuel particle takes to move about 63 percent of the way toward equilibrium moisture content with its environment. One hour fuels, fine grass and litter under one quarter inch, respond within an hour to temperature and humidity. Thousand hour fuels, 3 to 8 inch material, integrate weeks of weather and function as a drought signal.

Fine dead fuel moisture starts from reference fuel moisture, read at the intersection of temperature and relative humidity in the Fosberg tables, then corrected for month, time of day, slope, aspect, and shading. Rain, overnight humidity recovery, and cloud cover raise it. Their absence drops it.

How to prove the day was historically extreme

Energy release component is the available energy in British thermal units per square foot within the flaming front at the head of a fire, on an open ended relative scale: 24 is twice the potential heat release of 12. It composites all live and dead fuel moisture and deliberately omits wind and slope, which makes it stable enough for planning decisions 24 to 72 hours out and for year over year percentile comparison.

Energy release component at the 97th percentile for the date proves the day was historically extreme, which puts a reasonable operator on notice. Pull the station history from the archived daily fire danger records, not a summary.

The Haines index, and its limits

The Haines index, formally the lower atmospheric severity index, was developed in the 1980s to estimate the effect of atmospheric dryness and stability on wildfire growth. It sums a stability term, a temperature lapse rate between two pressure levels, and a moisture term, temperature minus dewpoint at a specified level, each scored 1 to 3. Three elevation variants exist: low uses 950 and 850 millibars, mid 850 and 700, high 700 and 500. The scale runs 2 to 3 very low, 4 low, 5 moderate, 6 high potential for large plume dominated growth.

Use it as corroborative context for plume dominated behavior, not as a causation element. It is a coarse index with known limitations, and an expert who leans on it will be made to concede that.

Red flag warnings are local, and there is no national threshold

A red flag warning tells land management agencies that red flag conditions are imminent or occurring, and issues when criteria will be met within 24 to 48 hours. A fire weather watch flags high potential for those conditions 12 to 72 hours out. Each local National Weather Service forecast office sets its own numeric criteria.

The Denver and Boulder office requires, for any three or more hours within a twelve hour period, either frequent gusts at or above 25 miles per hour together with relative humidity at or below 15 percent, or dry thunderstorms with 15 percent or greater coverage. Discretionary factors include wind shifts with frontal passage, the first heavy lightning after a long hot dry period, and overnight humidity recovery that stalls at or below 40 percent.

One wind, a different name in every state

Air is forced over a mountain range by a pressure gradient, descends the lee slope, and warms adiabatically at roughly 5.5 degrees Fahrenheit per 1,000 feet of descent while its absolute moisture content stays constant, so relative humidity collapses. The result is warm, extremely dry, fast air, often strongest at night when humidity would normally recover, which is why offshore wind events produce catastrophic fires.

The National Wildfire Coordinating Group names four elements of critical fire weather: low relative humidity, strong surface wind, unstable air, and drought. The upper ridge breakdown pattern is the classic three stage sequence, and the most dangerous western fire growth comes in stage two.

Rate of spread models are planning tools

Rate of spread is modeled from the 1972 Rothermel surface fire spread model, implemented in BehavePlus, FARSITE, FlamMap, and others. Inputs are fuel model, dead fuel moisture at the 1, 10, and 100 hour classes, live herbaceous and live woody moisture, midflame wind speed, and slope. FARSITE's own documentation lists the limitations: a surface fire model with empirical crown fire and spotting overlays, sensitive to fuel model selection and the midflame wind adjustment factor, and a planning tool rather than a forensic one. Published comparisons show real discrepancies among Rothermel based tools under varying wind and slope. Put that in the report before opposing counsel does.

Spotting, meaning firebrand lofting and downwind ignition of a receptive fuel bed, is modeled separately. A spot fire creates its own local origin with its own indicator field, which can be misread as the main origin, and for the reasons in chapter three every candidate origin has to be reconciled against the spotting hypothesis.

Where to get all of it, free, months later

Satellite hotspots bracket the time, they do not locate the origin

The geostationary refresh rate lets an investigator bracket an ignition window instead of only a day. State the limitations. Hotspot detections are not ignition points. They carry pixel scale positional uncertainty of 375 meters or a kilometer. Cloud, smoke, and canopy block them. The fire has to be large and hot enough at overpass. Their value is bracketing ignition time and showing direction and rate of early spread.

Chapter 6

Evidence and Spoliation

The physical evidence that establishes causation is usually owned by, located on, and removed by the defendant utility, often under genuine pressure to restore service. The utility is both custodian and target, which is why preservation letters, joint examination protocols, and spoliation motions are not optional.

The worked example: Edison and the re-energized lines

In mid January 2025, days after the Eaton Fire, plaintiffs' lawyers accused Southern California Edison of re-energizing the suspect transmission lines and destroying evidence. Edison answered that restoring customer power required it. Both can be true at once, and that is the problem.

Re-energization damages the physical record five ways. Fresh arcing at or near the same location overlays the original arc erosion. Metallurgical evidence of the first event, the annealing gradient, the oxide layer, the bead morphology, is altered by the second. The scatter field of solidified metal spatter gets a second, superimposed deposit. A conductor that was down is lifted, moved, or replaced. And the relay and supervisory control and data acquisition record now holds a second event that must be separated from the first before the first can be characterized.

That destroys the exact dataset an ignition opinion is built from, at the hands of the party whose ignition is in question, when litigation was plainly foreseeable. Whether it was sanctionable spoliation is for the court. Whether it should have been headed off with a preservation letter and a protocol on day two is not a close question.

What a preservation letter has to name

A generic demand to preserve "all relevant evidence" is worthless in a utility case. The records sit in named systems and the components carry span and asset numbers. Name them.

Physical components. The conductor, splice, connector, insulator, crossarm, pole, and every piece of hardware from the incident span and the two adjacent spans; the transformer and any protective device; and any component removed from the site, wherever it now sits.

Electronic operational records. Supervisory control and data acquisition time series. Relay event records and oscillography, including fault current magnitude, duration, and calculated distance to fault. Recloser and sectionalizer operation logs. Outage management system records with customer call timestamps. Advanced metering infrastructure last gasp data from meters downstream of the fault.

Decision records. Public safety power shutoff decisions for the event and the preceding week, and protection setting changes, including whether automatic reclosing was disabled and who decided.

Asset and vegetation history. Vegetation management inspection and work records for the span, with light detection and ranging scans and aerial patrol imagery; asset age, inspection, and maintenance history; and repeat fault history for the circuit.

Communications. All internal communications about the circuit, the span, the assets, and the decision to restore.

Joint evidence examination protocols, and why you negotiate one before anything moves

A written joint evidence examination protocol sets who may attend and on what notice, puts non destructive examination first, sets the photographic and video documentation required, states whether and how a component may be sectioned, cleaned, or metallurgically prepared, and fixes chain of custody and a retention rule for preserved samples. In California it is often negotiated among CAL FIRE, the California Public Utilities Commission Safety and Enforcement Division, the utility, and plaintiff and subrogation counsel before any component leaves the evidence locker.

After the October 2017 North Bay fires, Pacific Gas and Electric Company wrote the Safety and Enforcement Division on October 13, 2017, proposing that once CAL FIRE released each site, "the Safety and Enforcement Division and PG&E jointly visit each site," after which the division could "provide direction to PG&E on the particular evidence that it directs PG&E to retain." All of it was subject to the company's ability to restore service before removing operating equipment. That last clause is where the evidence goes.

The retention duty in California has a number attached

California Public Utilities Code section 316 and General Order 95, Rule 19 are the operative authorities. Decision 09-08-029 added Rule 19 on August 20, 2009, requiring full cooperation with Commission staff investigating any major accident or reportable incident, including immediate access to evidence, witnesses, and non privileged documents once the scene is secure and service restored. Evidence the utility collects in its own investigation "shall be preserved for at least five years."

Fire incident reporting is governed by Decision 14-02-015. A reportable fire is one that was self propagating and traveled more than one linear meter from the ignition point. Utilities file annual fire incident data reports on that basis, and those reports are public.

The federal spoliation standard, and what actually triggers the strong remedy

Federal Rule of Civil Procedure 37(e) governs electronically stored information in two tiers. On a finding of prejudice from the loss of information that should have been preserved and cannot be restored, the court may order measures no greater than needed to cure it. The strong remedies require a finding of intent to deprive another party of the information; the court may then presume the lost data was unfavorable, instruct the jury to presume it, dismiss, or enter default.

That intent finding is why the preservation letter is a litigation instrument, not a formality. A letter naming the systems and components, sent on a date certain, converts later loss from an accident into a decision.

Two rulings every wildfire lawyer should know

The Holiday Farm Fire and the Bonneville Power Administration. Judge Mustafa Kasubhai, D. Or., ruled on February 26, 2026 that the Bonneville Power Administration "willfully spoiled evidence because it moved trees of interest and allowed its contractors to materially disturb the ignition site without preserving evidence it knew it should, and could, preserve in reasonable anticipation of litigation." The fire ignited September 7, 2020. Before Forest Service investigators arrived on September 13, trees at the administration's right of way had been cut into pieces and pushed aside with heavy equipment. By September 20 a large tree lying on the transmission lines had been moved by unauthorized persons, and a planned three dimensional laser scan was never completed. The sanction was an adverse inference: the missing evidence is construed against the administration, and government counsel may not argue that the loss cuts against claimants. The docket number and official citation are in the punch list.

The Moonlight Fire, and the case where the investigators were the problem. Department of Forestry and Fire Protection v. Howell, Cal. Ct. App., Third Appellate District, December 6, 2017. The spoliation came from the fire agency's own origin and cause investigators, not the defendant. The trial court found CAL FIRE's lead investigator gave false deposition testimony about a white flag marking a point of origin, saying he could not explain a flag that photographs showed plainly, and found him not credible on the central issues. Investigators falsified interview summaries misrepresenting a witness's statements about rock strike causation, contradicted by the recorded transcript. The report omitted lookout observations and a witness's false alibi. The lead investigator destroyed his field notes after transferring them into the report, preventing cross examination when he later claimed memory lapses. The trial court imposed terminating sanctions and roughly $28 million in fees and expenses. The Court of Appeal reversed part of the fee award, holding it an abuse of discretion to award all defense expenses from the start of the case rather than only those caused by the discovery violations.

Chapter 7

Getting Your Expert Admitted

Wildfire origin and cause expertise is mostly experiential, which puts it inside Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999), not the safe middle of laboratory science. Add a validation study reporting a 103 degree mean directional error and a 2023 amendment rejecting the "goes to weight" answer, and admissibility becomes the most important motion practice in the file.

The framework, in four cases and one state variant

California also applies a gatekeeping duty to the reasons and matter underlying an opinion through Sargon Enterprises, Inc. v. University of Southern California, 55 Cal. 4th 747 (2012), and Evidence Code sections 801(b) and 802. That proposition sits in the verification punch list. Confirm it before you brief it.

What the 2023 amendment to Rule 702 changed, precisely

Effective December 1, 2023, Federal Rule of Evidence 702 lets a qualified expert testify if "the proponent demonstrates to the court that it is more likely than not that" the four listed conditions are satisfied. Subsection (d) now reads that "the expert's opinion reflects a reliable application of the principles and methods to the facts of the case." Two changes aim directly at forensic disciplines like fire investigation.

First, the explicit preponderance standard. The Committee Note says many courts had treated the sufficiency of an expert's basis, and the application of the methodology, as questions of weight rather than admissibility, and that this was an incorrect application of Rules 702 and 104(a). The proponent now carries the burden by a preponderance, at the threshold.

Second, the anti overstatement provision in 702(d). It targets forensic experts who claim more certainty than a subjective, error prone method supports.

Why this collides with the wildfire origin literature

Before December 2023, the standard defense of a flawed origin determination was that methodological deviations went to weight, not admissibility. That saved the government's expert in United States v. Idaho County Light and Power Cooperative Association, Inc., No. 3:17-cv-00391-CWD (D. Idaho, February 7, 2020), the best on point wildfire admissibility opinion available. A Forest Service captain applying the 2011 NFPA 921 was challenged on three grounds: failure to apply the methodology, no electrical systems expertise to opine on cause, and a cursory investigation with premature conclusions. The court noted the deviations but held they went to weight, citing Daubert's flexible standard and cross examination, and admitted her and the four retained experts who built on her work.

That reasoning is what the 2023 Committee Note disapproves. Meet any brief citing Idaho County Light and Power for a post December 2023 wildfire case with the Committee Note, and make the citing party explain why an opinion built on displaced reasoning still controls.

Then add the error rate. Your expert declares a specific origin area from directional fire pattern indicators, and the only published multi investigator validation study reports a 103 degree mean error against a 90 degree random baseline. The proponent must answer that by a preponderance, in the report. Do not hide from the study. Use the convergence position from chapter three, and note that the study measured single artifacts read in isolation, which is not what was done here.

The structure fire cases that set the baseline

Wildfire origin opinions are usually tested against the body of structure fire NFPA 921 rulings, and both sides use them.

Thompson is the model answer for an expert who departs from the guide. Deviation is survivable; unexplained deviation is not.

Arc mapping, and the attack you will face

NFPA 921 addresses arc mapping at section 9.11, with examination procedures at 9.11.7.3.2. The 2021 edition reclassified it as a fire pattern rather than an independent origin method, stressed survey completeness, and cautioned that arc sites are not necessarily at the area of origin, noting no apparent direct correlation between the origin and arcing at the closest geometric point.

The standard attack is Babrauskas, "Arc Mapping: New Science, or New Myth?", Fire and Materials 2017, reviewing the Carey compartment burn work, West and Reiter's 2005 finding of arcing diametrically opposite the origin, and Wheeler's 2015 finding that arcing localized to heavy fuel load rather than the origin. It puts the probability of a scientifically valid directional conclusion from solid core conductor arc mapping at about 0.7 percent, and names three asserted myths: that arc bead abundance indicates origin, that multiple arcs proceed upstream toward the source, and that internal appliance arcing near an exterior indicates external origin.

The law review treatment is May and Icove, "Arc Mapping Methodologies & the Pursuit of Magical Globules, Notches, & Beads: A Bridge Too Far to Establish Fire Origin?", 7 Lincoln Memorial University Law Review 38 (Fall 2020), collecting Glassman v. Home Depot USA, Inc. (arc mapping opinions shaky but admissible), Powell v. State Farm Fire & Casualty Co. (shifting hypotheses built on arc mapping), and Meemic Insurance Co. v. Hewlett-Packard Co. (visual observation insufficient without laboratory or metallurgical confirmation). The criticism is that visual and tactile examination cannot reliably tell arc damage from fire caused melting.

Your expert has to draw one distinction in advance. In wildfire cases, arc mapping usually means locating the arc or fault point along a circuit: read physical arc erosion on the conductor, add relay derived fault distance, and correlate to the flagged origin area. The structure fire critiques do not translate one for one. A cross examiner will use them anyway, and an expert drawing the distinction for the first time on the stand looks evasive.

How the circuits are reading amended Rule 702

The trend since 2024 is toward exclusion where the factual basis or the application is thin.

• EcoFactor, Inc. v. Google LLC, 137 F.4th 1333 (Fed. Cir. 2025). A twenty million dollar verdict reversed; opinions unreliable and therefore inadmissible for insufficient factual basis under 702(b).
• Sprafka v. Medical Device Business Services, 139 F.4th 656 (8th Cir. 2025). Opinions lack reliability where the factual basis is inadequate; an express departure from the circuit's prior liberal admission posture.
• Nairne v. Landry, 2025 WL 2355524 (5th Cir. Aug. 14, 2025). The proponent must show admissibility more likely than not.
• Harris v. FedEx Corporate Services, Inc., 92 F.4th 286 (5th Cir. 2024). A trial court abdicates its role as gatekeeper by admitting without a proper 702(b) foundation.
• Engilis v. Monsanto Co., 2025 WL 2315898 (9th Cir. Aug. 12, 2025). There is no presumption of admission; challenges go to weight only if the court first finds a sufficient basis.
• Bulone v. Monsanto Co., 2025 WL 2730843 (9th Cir. Sept. 25, 2025). Reliability applies to the expert's application of the methods.
• In re Onglyza (Saxagliptin), 93 F.4th 339 (6th Cir. Feb. 13, 2024). District courts have an independent duty; exclusion proper for cherry picked data.
• Baker v. Blackhawk Mining, LLC, 141 F.4th 760 (6th Cir. 2025). Causation opinions lacking reliable factual bases are excludable under 702(b) and (d).
• Davis v. Sig Sauer, Inc., 126 F.4th 1213 (6th Cir. 2025). Opinions failing to reasonably apply principles or methodologies to underlying facts warrant exclusion.

The contrary authority worth knowing is Sommerville v. Union Carbide Corp., 149 F.4th 408 (4th Cir. 2025), where the majority overlooked the 2023 amendments and held that factual questions go to weight and credibility rather than admissibility, over a dissent invoking Rule 702(b). It is useful to a proponent and vulnerable on appeal. The pre amendment rule being displaced is Loudermill v. Dow Chemical Co., 863 F.2d 566 (8th Cir. 1988), holding that the factual basis of an expert opinion goes to credibility rather than admissibility.

Chapter 8

Utility Liability

California is the outlier. Everyone else runs a variation on negligence. The difference is the highest leverage doctrine in this practice area, because it decides whether the plaintiff must prove anything at all about the defendant's conduct.

California is very unusual in that normally this doctrine only applies to government entities. The plaintiffs won't have to prove [the utility] poorly maintained these transmission wires or did a shoddy job.

Daniel Farber, Professor, University of California Berkeley School of Law, Reuters, January 15, 2025

Inverse condemnation, and why it operates as strict liability

Inverse condemnation derives from Article I, section 19 of the California Constitution: private property may not be taken or damaged for public use without just compensation. Where a public improvement damages private property, the entity pays regardless of fault.

City of Oroville v. Superior Court (2019) 7 Cal.5th 1091 controls: "If damage to private property is substantially caused by the inherent risks of the design or construction of a public improvement, a public entity must provide just compensation for the damage, whether it was intentional or the result of negligence by the public entity."

No duty, breach, or negligence showing is required. The plaintiff must establish only that the utility's equipment was a substantial cause of the fire. Attorneys' fees and costs are recoverable under Code of Civil Procedure section 1036, which negligence claims do not provide, and that fee provision is one of the largest drivers of settlement value in California wildfire litigation.

The cases that put an investor owned utility inside the doctrine

The Barham facts are the archetype. Santa Ana winds caused Edison's 12,000 volt line to contact a 33,000 volt line, and a failed lightning arrester ignited the Mill Creek fire. The court held that Article I, section 19 compensates any actual physical injury to real property proximately caused by a public improvement as deliberately designed and constructed, and that a circuit serving more than a thousand households established the public benefit. It distinguished Cantu v. Pacific Gas & Electric Co., which involved individual service lines, from permanent transmission towers built to carry electricity across a much wider area.

The utilities' rejoinder, and why courts keep rejecting it

The recurring argument is that unlike a municipality, an investor owned utility cannot unilaterally raise rates to spread the loss, so the cost spreading rationale fails. Courts reject it. The Simple Avo Paradise Ranch court: "the government has chosen to grant a franchise and delegate the furnishing of electricity to Edison rather than operating the utility itself. Such a delegation does not remove the policy justifications underlying inverse condemnation liability."

This court is not tasked to determine what the law should be and is merely tasked with interpreting what the law is and has been for 125 years.

Hon. Dennis Montali, United States Bankruptcy Judge, Northern District of California, rejecting Pacific Gas and Electric Company's attempt to escape inverse condemnation, Courthouse News Service, November 27, 2019

The live frontier: is an idle line a public improvement

Southern California Edison argues in the Eaton Fire litigation that a transmission line idle since 1971 and serving no customers is not public use infrastructure and so does not trigger inverse condemnation. Judge Laura Seigle of Los Angeles Superior Court set a hearing for August 11, 2026. It is the first serious limiting argument since Oroville, and if it succeeds it changes the value of every case involving a decommissioned or standby asset.

Negligence per se hooks in California

Even where inverse condemnation is available, plead negligence: the fee provision does not cover everything, and a jury finding of unreasonable conduct changes settlement dynamics. Three regulatory sources supply the standard of care.

Public Resources Code section 4292 requires clearance around poles and towers supporting specified equipment in hazardous fire areas. Section 4293 sets minimum radial clearances between conductors and vegetation, scaled by voltage. General Order 95 governs overhead line construction, including clearances, hardware, inspection, and Rule 19's evidence retention duty. Judge Alsup made full compliance with 4292 and 4293 an express condition of Pacific Gas and Electric Company's federal criminal probation.

Negligence per se requires that the injury result from an occurrence the statute was designed to prevent and that the plaintiff be in the protected class. Both are usually satisfied in a wildfire case, and both still have to be pleaded. A federal court dismissed negligence per se claims in United States v. PacifiCorp (D. Or., Aug. 26, 2025) while letting breach of license claims resting on the utility's Federal Energy Regulatory Commission obligations proceed.

How the other states differ

Oregon: pure negligence, no statutory shield, and the largest verdict exposure in the country. PacifiCorp's Labor Day 2020 exposure rests on findings of negligence, gross negligence, recklessness, and willful conduct, not strict liability. In James v. PacifiCorp, Multnomah County Circuit Court No. 20CV33885, Judge Steffan Alexander, a Phase I jury in June 2023 found the utility liable for four fires, awarded more than $70 million to seventeen named plaintiffs, and set punitive damages at 0.25 times economic and non economic damages, a one to four ratio well below the constitutional single digit guidepost. Oregon also doubles economic damages under its trespass and timber statutes. Across the published Phase II trials, PacifiCorp's own accounting shows $1.2307 billion in net damages to 190 plaintiffs: $130.15 million doubled economic, $895.66 million non economic, and $240.19 million punitive, less $35.33 million in insurance credits.

On April 8, 2026 the Oregon Court of Appeals reversed and remanded in James v. PacifiCorp, No. A183140, holding the trial court erred by instructing the jury it could "assume that the evidence at the trial applies to all class members," when "much of the evidence introduced during the Phase I trial related to particular issues concerning particular wildfires." The instruction let jurors carry evidence about specific fires over to class members harmed by different fires more than a hundred miles apart. The Oregon Supreme Court granted review. The holding is about causation and individuation, and it changes how origin evidence is presented in any multi fire class action.

Colorado: negligence, no cap, no presumption statute. The Marshall Fire litigation proceeded on ordinary tort theories against Xcel Energy, with causation genuinely disputed. Colorado's activity has been regulatory rather than liability reform. No Colorado statute limiting utility wildfire liability was found in the research for this book, and that negative finding is in the punch list.

Hawaii: negligence, with a judicially imposed insurer subrogation bar. No strict liability. The Maui Fire Cases ruling removed subrogation carriers as independent claimants. Act 258 of 2025 directs the Public Utilities Commission to study a liability cap and a recovery fund, with a decision expected in early 2027 and no cap amount yet proposed.

Texas: negligence with a statutory safe harbor. House Bill 145 of 2025 requires wildfire mitigation plans and bars liability where the utility complied with its plan and was not negligent. The Texas Attorney General pleaded around it in Texas v. Xcel Energy with negligence, gross negligence, trespass, public nuisance, unjust enrichment, violations of the Public Utility Regulatory Act and Commission rules, and Deceptive Trade Practices Act violations. That breadth is deliberate, and it is a template.

The 2025 legislative wave, and the architecture it shares

Seven states enacted utility favorable liability legislation in 2025, all sharing one architecture: an approved mitigation plan converts into a negligence presumption or a safe harbor. Oregon has the largest active verdict exposure in the country and passed nothing.

That put the fear of God in all the states across the West.

Michael Wara, Stanford University, on the effect of the PacifiCorp verdicts on Western legislatures, The Spokesman-Review, March 1, 2026

Chapter 9

Who Actually Pays

California decoupled liability from payment on purpose. A utility can be strictly liable under chapter eight's inverse condemnation rules and still book a net after tax charge to shareholders of nine million dollars. That is Assembly Bill 1054 working as designed, on top of ratepayer funded self insurance.

The decision that broke the old model

San Diego Gas & Electric paid about $2.4 billion in settlements, costs, and legal fees for the 2007 Witch Creek, Guejito, and Rice fires, with $1.1 billion from liability insurance and $824 million from third party settlements. It sought $379 million from its wildfire expense memorandum account. Decision 17-11-033, November 30, 2017, denied the request in full.

The Commission applied the reasonable and prudent manager standard: "The term reasonable and prudent means that at a particular time any of the practices, methods and acts engaged in by a utility follows the exercise of reasonable judgment in light of the facts known or which should have been known at the time the decision was made." The utility failed its burden on each fire. On the Witch Fire, the response to four faults over six and a half hours was unreasonable; it had known since 2001 that automatic reclosers increase fire risk and did not act, despite red flag warnings and fires nearby. On the Guejito Fire, a 3.3 foot clearance violation persisted six years despite inspections. On the Rice Fire, it deviated from annual trimming for a specific tree and failed to designate it a reliability tree.

The utility called that an unconstitutional taking: strict liability without fault, no cost recovery. The California Supreme Court declined review and the United States Supreme Court denied certiorari on October 7, 2019. That gap drove Pacific Gas and Electric Company into bankruptcy and produced Assembly Bill 1054.

What Assembly Bill 1054 actually did

Enacted July 12, 2019 as an urgency statute, it created a wildfire fund of about $21 billion from three sources: $10.5 billion from extending an existing $2.50 per month customer charge, $7.5 billion in initial shareholder contributions from Pacific Gas and Electric Company, Southern California Edison, and San Diego Gas & Electric, and roughly $3 billion over ten years from ongoing shareholder contributions of $300 million a year. Utility contributions come from shareholders alone and are not recoverable from ratepayers.

The core innovation is a burden shift. Ordinarily a utility must prove its conduct reasonable to get cost recovery. With a valid safety certification, other parties must first show "a serious doubt as to the reasonableness" of that conduct before the utility has to dispel it. Certification requires a board level safety committee, implementation of the utility's most recent safety culture assessment findings, and an executive compensation structure built to promote safety, among other conditions.

Reimbursement of unreasonable costs is capped, with two exceptions: conduct involving "conscious or willful disregard of the rights and safety of others," and failure to maintain a valid safety certification.

San Diego Gas & Electric recovered zero percent under the pre-1054 standard. Southern California Edison settled its cost recovery application on September 19, 2025 and may recover $2 billion for the 2017 and 2018 events: $1.6 billion of uninsured claims payments and $400 million of legal costs through May 31, 2025, plus 35 percent of losses paid after. With prior approvals, its expected recovery is 43 percent of $5.6 billion in losses. Zero to forty three percent, same regulator, different statute.

The arithmetic that explains the whole system

As of June 30, 2026, Southern California Edison had recorded $1.6 billion in losses on Eaton Fire settlements, against $917 million in self insurance recovery, $645 million in Wildfire Fund recovery, and $70 million through Federal Energy Regulatory Commission rates. The net after tax charge to earnings, at that date and each prior reporting date, was $9 million, reflecting a $12.5 million shareholder contribution.

Self insurance here is funded through rates. The Wildfire Fund is capitalized from a customer charge and shareholder contributions. Federal Energy Regulatory Commission rates are paid by transmission customers. Trace each arrow and the money ends where it started, with the people who buy electricity, including those whose houses burned.

Compare Colorado, where shareholders actually paid

Xcel Energy settled the Marshall Fire litigation for $640 million, roughly $353 million of it recoverable from insurers. A non recurring $287 million charge hit shareholders in the third quarter of 2025, plus $12 million in the fourth for estimated remaining settlement and legal costs. Colorado has no wildfire fund, no ratepayer funded self insurance of the California kind, and no statutory shield. When the insurance ran out, the loss landed on the equity.

Xcel says so in its own annual report risk factors. "While we carry liability insurance, given an extreme event, damage amounts could exceed our coverage (as experienced with the Marshall Wildfire settlement in 2025) and negatively impact our results of operations, financial condition or cash flows."

What the victim receives, in cents

Settlement totals are not what claimants get. The Pacific Gas and Electric Company Fire Victim Trust, established July 1, 2020, was funded with $5.4 billion in cash plus company stock, and supplemental cash of $758 million in January 2021 and $592 million in January 2022, covering the 2015 Butte Fire, the twenty two major 2017 North Bay fires, and the 2018 Camp Fire. The pro rata payment percentage rose from 30 percent in March 2021 to 45 in February 2022, 60 in January 2023, 66 in March 2024, and 70 in September 2024, where it remains. Aggregate determined award value hit $19 billion as of December 26, 2023, later reported at $19.43 billion, against $13.5 billion of funding. Total paid was $13.71 billion as of June 30, 2026, with 71,787 claimants having received determination notices. The final pro rata true up is still unpaid, nearly six years on.

Maui is worse in a different way. The $4.037 billion global settlement pays in four annual tranches of roughly a billion dollars across 21,750 claimants and 94,816 unique claims, averaging about $190,000 per victim, or $47,000 a year. Insurers who had paid $3.03 billion were projected to recover about $216 million.

The settlement was done in record time, and today's March 27th, 2026, and no one's seen a penny.

Hon. Peter Cahill, Maui Circuit Court, hearing of March 27, 2026, Honolulu Civil Beat, April 28, 2026

What is coming

Senate Bill 254, signed September 2025 as an urgency statute, adds about $18 billion to the fund, authorizes up to $9 billion in bonds, allows another $3.9 billion in collections if needed, extends the ratepayer charge through 2045, creates a continuation account for new liabilities, and required a study of alternative structures to socialize catastrophe risk.

The California Public Utilities Commission reported on January 30, 2026, identifying "potential changes to utility wildfire liability standards, including modifications to the inverse condemnation strict liability framework and/or the capping or limitation of utility tort liability," and recommending a wider pool of fund contributors, possibly including local governments. The California Earthquake Authority, which administers the fund, goes further: amend the state Constitution to move from strict liability to fault based negligence, cap additional living expenses at five years, eliminate punitive damages against investor owned utilities, remove insurance subrogation, and limit attorneys' fee recovery. Removing subrogation alone is estimated to cut utility payments by 35 to 40 percent. No reform bill had been introduced when this went to press, and the session was closing. Check the status before relying on the current regime.

Chapter 10

Criminal Exposure

Five North State District Attorneys said criminal penalties would have capped out at $329,417 had they prosecuted Pacific Gas and Electric Company for the Dixie Fire. They went civil instead. On April 11, 2022 the company paid $29.5 million to nonprofits and $1.25 million to District Attorney offices, and took a safety monitor costing up to $15 million a year for five years.

Corporate criminal fines are trivial next to the harm. The leverage is restitution, monitorship, and injunctive commitments, and prosecutors get all three without the risk of trial.

The Camp Fire plea, and what a maximum penalty looks like

A Butte County grand jury of nineteen, sworn March 25, 2019, heard almost a hundred witnesses and about 1,600 exhibits across roughly 6,000 pages of transcript, then indicted Pacific Gas and Electric Company on one felony count of unlawfully and recklessly causing the Camp Fire through gross negligence in maintaining its power line, 84 counts of involuntary manslaughter, and three special allegations: great bodily injury to a firefighter, great bodily injury to more than one surviving victim, and causing approximately 18,804 structures to burn.

A C hook suspension device failed on transmission Tower 27/222, letting an energized line arc against the steel tower, and molten metal ignited dry brush. The hook had roughly 97 years of rotational wear, deteriorating about fourteen sixteenths of an inch before it fractured.

The company pleaded guilty to all counts on June 16, 2020 and was sentenced June 18 to a $3.5 million criminal fine, the stated maximum for corporate crimes, plus $500,000 to the District Attorney for investigation costs and up to $15 million over five years for canal restoration.

If these crimes were attributed to an actual human person rather than a corporation, the applicable sentence would be 90 years to be served in state prison.

Hon. Michael Deems, Butte County Superior Court, at the Camp Fire sentencing, Courthouse News Service, June 18, 2020

The company then sought to deduct the $4 million from the $13.5 billion Fire Victim Trust, and the District Attorney and victims objected.

San Bruno, and the probation that became a wildfire supervision order

Pacific Gas and Electric Company was convicted of six federal felonies from the San Bruno pipeline explosion: five willful violations of the Natural Gas Pipeline Safety Act of 1968 and one count of obstructing an agency proceeding under 18 United States Code section 1505. On January 26, 2017, Judge Thelton E. Henderson imposed the $3 million statutory maximum fine, five years of probation, a corporate compliance and ethics monitorship, a compliance program due within six months, $3 million in public advertising, and 10,000 hours of community service.

Judge William Alsup turned the probation into a wildfire safety instrument. In March 2019 he proposed full compliance with Public Resources Code sections 4292 and 4293, mitigation plan vegetation targets, unannounced monitor inspections, monthly vegetation reporting, a dividend ban until compliance, and de-energization during extreme fire danger. His April 2, 2019 order barred dividends and redirected the funds to wildfire risk reduction. His April 29, 2020 order made in house tree inspectors double check contract trimmers, made the company catalog the age and condition of every piece of hardware on every transmission tower and line and videotape every inspection, and made inspection contractors carry insurance "to cover losses suffered by the public should their inspections be deficient and thereby start a wildfire."

Probation expired January 25, 2022. During the five year term more than thirty wildfires were blamed on the company, destroying more than 23,000 homes and businesses and killing more than a hundred people.

In these five years, PG&E has gone on a crime spree and will emerge from probation as a continuing menace to California. We have tried hard to rehabilitate PG&E. As the supervising district judge, however, I must acknowledge failure.

Hon. William Alsup, on the expiration of probation, National Public Radio, January 24, 2022

Zogg, and why the prosecution collapsed

Shasta County District Attorney Stephanie Bridgett charged the company on September 24, 2021 with 31 counts, eleven felonies and twenty misdemeanors, including involuntary manslaughter for four deaths, over a gray pine marked for removal in 2018 that fell onto a power line. Twenty counts were dismissed at the preliminary hearing in early 2023, and on May 31, 2023 the rest went under a $50 million settlement: $45 million to community and nonprofit organizations, a $5 million civil penalty, no customer cost recovery, and a five year independent monitorship.

The court found the People "failed to identify a standard of care" and that there was "no evidence that a risk involving the tree was visible before the fire." A corporate wildfire prosecution requires proof of a standard of care and a visible, knowable risk, which is far harder than proving that equipment started a fire.

This decision has severely limited our ability to accomplish the goal that we wanted through criminal prosecution.

Stephanie Bridgett, Shasta County District Attorney, after the Zogg Fire dismissal, KTVU, May 31, 2023

The civil judgment in lieu of prosecution

Sonoma County filed 33 criminal counts over the 2019 Kincade Fire in April 2021 and dismissed them in April 2022 for a $20.25 million civil judgment: $6 million to local nonprofits, $6 million to Santa Rosa Junior College, $7.5 million in civil penalties, and $750,000 in investigative and prosecutorial costs. The company also added 80 wildfire safety positions in Sonoma County and took five years of independent monitoring.

Although criminal charges are dismissed, the level of punishment and oversight provided by this Judgment is greater than could be achieved against a corporation in criminal court.

Jill Ravitch, Sonoma County District Attorney, April 2022

Set against the Dixie cap, the pattern is complete. Prosecutors trade the conviction for money and control because in dollars the conviction is worth almost nothing.

Where other utilities landed

Edison's securities filings say it "is not aware of any basis for felony liability with regards to the Eaton Fire," warn it "could be subject to material fines, penalties, or restitution," and state that "any fines and penalties incurred in connection with the Eaton Fire will not be recoverable from insurance, from the Wildfire Fund, or through electric rates." That last item is the one wildfire cost that lands on shareholders with no way out.

We make our decisions based on evidence, and not based on emotion. If we were to tell you today that we were filing charges it would be wrong and unethical.

Michael Dougherty, Boulder County District Attorney, declining to charge over the Marshall Fire, June 8, 2023

Individuals: arson, reckless, negligent

California Penal Code section 452, unlawfully causing a fire, covers recklessly setting fire to any structure, forest land, or property, and was amended effective January 1, 2025. Penalties tier by what burned. Great bodily injury: two, four, or six years. Inhabited structure or property: two, three, or four years. Structure or forest land: sixteen months, two, or three years. Property only: a misdemeanor. Willful and malicious arson is section 451, and the distinction is malice.

Colorado runs a two track statute. Intentional wildfire under Colorado Revised Statutes 18-13-109.5 produced fifteen cases statewide over a decade; reckless or negligent setting fire to woods or prairie under 18-13-109 produced 261 in the same period.

Federal hooks are 18 United States Code section 844(f)(1), destruction of property by means of fire; section 844(i), arson affecting interstate commerce; and section 1855, timber set afire.

Five individual prosecutions worth knowing

The Palisades Fire. United States v. Jonathan Rinderknecht, No. 2:25-CR-833-AH (C.D. Cal.), indicted October 15, 2025 on three counts under sections 844(f)(1), 844(i), and 1855, maximum exposure 45 years. Mistrial June 26, 2026 on a ten to two deadlock for acquittal. Retrial October 19, 2026, defendant detained.

The Line Fire. People v. Justin Halstenberg, convicted May 2025 and sentenced October 1, 2025 to sixteen years to life for igniting both the Bacon and Line Fires in September 2024. The Line Fire burned about 44,000 acres and injured six firefighters.

The Old Fire. People v. Rickie Lee Fowler, the high water mark of individual wildfire liability: convicted August 2012 of five counts of first degree murder and two counts of arson, sentenced to death January 28, 2013. The five victims died of heart attacks during the fire or while evacuating.

The Caldor Fire, and the case that collapsed. Two defendants arrested in December 2021 and charged with reckless arson under section 452. In early January 2024, El Dorado County Superior Court Judge Vicki Ashworth held the evidence did not meet the recklessness standard, which requires a defendant who "knows their actions present a substantial and unjustifiable risk but consciously disregards that risk." The fire charges were dismissed and firearms counts survived. Recklessness has a subjective awareness element, and that is where these cases fail.

The El Dorado Fire, the gender reveal fire. People v. Jimenez. One defendant pleaded to felony involuntary manslaughter for a firefighter's death plus two felony counts of recklessly causing fire to an inhabited structure, and was sentenced February 9, 2024 to one year county jail, two years felony probation, and 200 hours of community service. The other pleaded to three misdemeanors and got one year of probation plus community service. Joint restitution was $1,789,972.

Restitution against people who can never pay

Under California Penal Code section 1202.4(a), a crime victim gets restitution directly from the defendant convicted of it, and subdivision (f) requires full restitution for every determined economic loss. Subdivision (j) coordinates the two systems: "Restitution collected pursuant to this subdivision shall be credited to any other judgments for the same losses obtained against the defendant arising out of the crime for which the defendant was convicted." Restitution does not bar the civil suit. It offsets it.

Restitution against individuals routinely exceeds any realistic ability to pay, because it is set by suppression cost accounting rather than means.

• Eagle Creek Fire, Oregon, 2017. A fifteen year old pleaded guilty to starting a roughly 47,000 acre fire with a smoke bomb. On May 21, 2018, Judge John Olson ordered $36.6 million in restitution, $21.1 million of it to the Forest Service and $12.5 million to the Oregon Department of Transportation, plus five years of probation and 1,900 hours of community service.
• Hayman Fire, Colorado, 2002. A Forest Service employee received six years in federal prison and $14.6 million in federal restitution. Her state sentence, originally twelve years and overturned in 2004, became fifteen years of probation and 1,500 hours of community service on resentencing in March 2008, with state restitution reported at $44.5 million.
• Mendocino National Forest, 2003. A twenty six year old pleaded to a federal misdemeanor of leaving a fire burning or unattended and was ordered to pay $18.2 million in restitution plus a five year ban from the forest. The fire ran 6,058 acres and cost $33 million to suppress.

The restitution bears a sufficient relationship to the gravity of the offenses and does not shock the moral sense of reasonable people.

Hon. John Olson, ordering $36.6 million against a fifteen year old, Oregon Public Broadcasting, May 21, 2018. Defense attorney Jack Morris called the request "absurd" for a fifteen year old

Federal cost recovery is now the largest number in the room

The federal damages formula runs the same way across the settlements: suppression costs incurred, natural resource damages and restoration costs for burned federal land, loss of use and public access harms including trail and campground closures, habitat destruction, protected species and cultural resource impacts, and sometimes land conveyed in kind.

It is our intention that Southern California Edison, the company itself, and not the rate payers will bear the burden of these costs.

Bill Essayli, federal prosecutor, Central District of California, CBS Los Angeles, September 4, 2025

California has a parallel state mechanism in Health and Safety Code section 13009. A person who negligently sets a fire, or fails to correct a fire hazard, is liable for suppression costs and for rescue or emergency medical services. Those costs are a debt, collectible like a contract debt, and several participating public agencies may name one agency to sue for all of them.

The prescribed burn problem

Liability for an escaped prescribed burn splits four ways across the states. A 2006 survey classified eighteen states on simple negligence, four on gross negligence, six on strict liability, and twenty two with no clear statutory standard. That survey is stale and is in the punch list. Re-survey your state before advising anyone.

California Civil Code section 3333.8, from Senate Bill 332 of 2021, limits prescribed burning liability to gross negligence while keeping exposure open at the top end: "This section shall not be construed to grant immunity from fire suppression or other costs to any person whose conduct constitutes gross negligence." Protection carries seven conditions. The burn must be for wildland fire hazard reduction, ecological maintenance and restoration, cultural burning, silviculture, or agriculture. A certified burn boss must approve a written prescription with risk mitigation measures, and the crew must comply with it. The burn must be authorized under the Public Resources Code. The landowner or tribe must permit it. Air quality permits must be in order. Cultural burns by a cultural fire practitioner are exempt from the burn boss approval and prescription compliance requirements.

Hermit's Peak and Calf Canyon. Two federal ignitions merged into the largest wildfire in New Mexico history. The Forest Service ignited the Las Dispensas and Gallinas prescribed burn on April 6, 2022; it escaped the unit boundary and was declared a wildfire the same day, roughly four hours after a successful test fire at 12:34. The Calf Canyon Fire began April 19, 2022, traced to a January 2022 pile burn that had stayed dormant under the surface.

The Forest Service after action review shows a plan failing, not a person. The plan did not account for post thinning fuel model changes or heavy fuel concentrations near firelines. Relative humidity "dropped well below the forecasted range of 9-13%, reaching 6%," past prescription minimums, and crews took weather observations hourly even after it fell below the 12 percent threshold that should have triggered more frequent monitoring. "Suppression resources (holding and contingency) were not calculated adequately." Foliar moisture trended down far enough to enable crown fire. The review concluded that personnel "followed their approved prescribed fire plan" and "believed they were within the approved prescription limits."

Congress responded with the Hermit's Peak/Calf Canyon Fire Assistance Act, Public Law 117-180, September 30, 2022, supplemented by Public Law 117-328 on December 29, 2022, appropriating $3.95 billion. Implementing regulations sit at 44 Code of Federal Regulations Part 296. Payout reached $3.51 billion across 24,502 claims as of June 23, 2026. The federal government paid for its own escaped burn through a claims program instead of tort litigation, and any future federal escape gets measured against that.

Chapter 11

The Marshall Fire

I watched this fire from my window. Friends of mine lost their homes. I tried a wildfire property loss case in Boulder County.

Twelve hours, and the most destructive fire in Colorado history

Fifteen minutes from the first 911 call to out of control. Thirty five minutes to an evacuation order. Fifty one minutes to the fire arriving in a town three miles away. No fire agency in the country is resourced to beat that clock.

The conditions were a two year setup

Peak gusts reached 115 miles per hour, driven by a mountain wave event. The instrument and station location for that reading are unverified and in the punch list.

A wet spring in 2021 produced above average grass growth, a warm dry summer and fall cured it into continuous fine fuel, and no significant snow had fallen that winter. A wet year followed by a dry year is not a mitigating circumstance. It is a loading mechanism: the wet year builds the fuel and the dry year primes it.

The dual ignition finding

The Boulder County Sheriff's Office announced the results of an eighteen month investigation, case 21-6301, on June 8, 2023: "Burn patterns, video footage, and satellite imagery show that the Marshall Fire was comprised of two fires that eventually merged becoming the Marshall Fire."

Ignition one, a residential slash burn. From the release: "It was determined that one fire started on a residential property located at 5325 Eldorado Springs Drive where a legal burn had occurred the previous week and had been covered with dirt but had been re-ignited due to the high winds on December 30." The burn was lit December 24, 2021. A community member "allowed the fire to burn to coals at which point he covered the fire with dirt, but did not extinguish the coals with water." Local firefighters were notified and reportedly raised no concern. No charges followed, and slash burning was not illegal in Boulder County.

That is the most common failure in debris burning cases. Covering coals with dirt insulates them; it does not put them out. Water does. Six days later a 100 mile per hour wind stripped the dirt off and put oxygen back.

Ignition two, a power line. "The second fire was started by arcing powerlines located just south of the Marshall Mesa Trailhead, located southeast of the intersection of Highway 93 and Eldorado Springs Drive." Boulder County's later account says investigators concluded the fire was "most probably caused by hot particles discharged from Xcel Energy powerlines." The evidence chain had three links: trail camera footage showing a detached power line with smoke and flames nearby, inspection by United States Forest Service experts, and engineering analysis confirming the line discharged hot particles.

Coal seams, excluded but not eliminated. "Investigators found no evidence that elevated underground fire activity in the coal seams produced surface temperatures capable of starting a fire, although that cannot be completely ruled out." That is how certainty language should read: an affirmative finding of no evidence, no claim of absolute exclusion.

No charges. "Applying this standard, no criminal charges will be filed related to the ignition of the Marshall Fire. Though the fire devastated thousands of people, there is insufficient evidence that a crime was committed." Xcel Energy publicly disputed the power line conclusion the same day.

The losses, and a denominator discrepancy you should handle openly

Boulder County's recovery program page uses 1,109 homes destroyed as the rebuild tracking denominator, not 1,084. Both are in circulation. Pick one, state which, and footnote the other. Do not average them or switch between them inside one analysis.

Two deaths. About 6,200 acres. More than $2 billion in damage. The most destructive and costliest wildfire in Colorado history, surpassing the 2013 Black Forest Fire within twelve hours.

The litigation and the settlement

Public Service Company of Colorado, the Xcel subsidiary, disclosed the numbers in securities filings: as of December 31, 2025, 307 complaints with 4,087 plaintiffs, $640 million in Marshall settlements, $353 million in insurance settlements receivable, and $5 million in current settlement liabilities.

The shareholder share came in pieces: a $35 million charge in fiscal 2023, nothing in fiscal 2024, a non recurring $287 million charge in the third quarter of 2025 for the settlement, and another $12 million in the fourth quarter for estimated remaining settlement, legal, and other costs.

The judge canceled the trial on September 24, 2025, one day before it was to begin, after the parties reported a settlement in principle, and nearly all of the roughly 4,000 plaintiffs signed. As of January 13, 2026 only four plaintiffs had open claims or unsigned offers, and roughly 600 minors required separate court approval before a separately appointed judge. Individual amounts are confidential.

Despite our conviction that [Public Service Co.] equipment did not cause the Marshall Fire or plaintiffs' damages, we have always been open to a resolution that properly accounts for the strong defenses we have to these claims.

Bob Frenzel, Chairman, President and Chief Executive Officer, Xcel Energy, Colorado Sun, September 24, 2025

The court, case caption, case number, and judge for the consolidated proceeding, whether a class was certified, and the theories pleaded are all unverified and sit in the punch list. The Twelve Tribes was not named as a defendant.

The insurance aftermath

United Policyholders reported that 74 percent of homeowners who filed claims after the Marshall Fire were underinsured by an average of $139,000. Pin that pairing to its underlying study before citing it. The same organization elsewhere says roughly two thirds of the Colorado wildfire survivors it works with report underinsurance, and its twelve month survey of 574 households found 80 percent with not enough insurance and an average shortfall of $110 per square foot.

Douglas and Amanda Mayfield of Louisville bought $419,000 in dwelling coverage in 2021 on their agent's recommendation. Their insurer's rebuild estimate was $555,055. The actual rebuild cost was $850,000. Sarah Sierra, 76, of Superior was insured for $120,000 with a rebuild estimated at roughly $300,000 more.

The agents are beholden to the software.

Shirin Chahal, attorney for the Mayfields, August 15, 2022

Four years on

No official figure exists for how many families never returned, so build it from the table. At least 178 of 1,109 destroyed properties, 16 percent, show no completed path back: 85 with no activity, 49 sold without permits, 44 pending or listed. Present that as a floor, not a count.

Boulder County's other programs show the administrative load that follows a fire. Debris removal cost $35 million and served 566 property owners, a use tax rebate distributed over $2.1 million to 676 households, and navigation services logged more than 1,200 appointments.

I grabbed our dog and two coats on the way out of the house. In the car, we realized I was the only one with shoes on, my wife hadn't even had time to grab her purse, and the two coats I got were both mine, which was not helpful.

Jon Solove, Superior resident, 5280 magazine, July 5, 2022

Chapter 12

Eaton and Palisades

Everyone was warned. The forecast was right, went out over and over for a week in escalating language, and named the outcome that happened.

The warning sequence

The National Weather Service Los Angeles office predicted a life threatening windstorm that would "accelerate to dangerous levels," the "most destructive windstorm seen since 2011," producing "widespread power outages and downed trees."

The antecedent conditions supported the forecast. El Niño had flipped to La Niña by December 2024, Southern California recorded the driest start to a rainy season on record and the driest nine month period on record before the windstorm, and most of Los Angeles County was in moderate drought by late December. Peak gusts on the day: 100 miles per hour at Mount Lukens Truck Trail, 98 in the Santa Monica Mountains, 84 at Magic Mountain Truck Trail, 62 at Escondido Canyon, 55 at Van Nuys Airport.

The two fires

At least 31 official fatalities, more than 18,000 homes and structures destroyed, more than 57,529 acres. An August 2025 study linked up to 440 excess deaths to the fires; attribute that figure to the study by name rather than folding it into the official count. CAL FIRE's Palisades page lists a start time of 2:30 a.m. on January 7, not the widely reported 10:30 a.m. Do not print either until that conflict is resolved.

The loss estimates diverge by a factor of five

Insured loss and total economic loss are different quantities, and total economic estimates vary by more than a factor of five depending on what the modeler counts. Name the source and definition every time. Never present a single number as the loss.

The California Department of Insurance claims tracker reported totals as of February 5, 2025: 33,717 home, business, and living expense claims filed; 5,597 auto claims; $6.94 billion paid on residential and commercial claims; $73 million paid on auto; and 19,854 claims partially paid under advance payment laws.

The water ran out

The Santa Ynez Reservoir, capacity 117 million gallons, was empty on January 7, offline since February 2024 for cover repairs due by April 2024, and officials acknowledged that higher water pressure would have been available had it been in service. Three one million gallon Pacific Palisades tanks were exhausted by early on January 8, and the chief executive of the Los Angeles Department of Water and Power reported all fire hydrants dry by approximately 3 a.m. Demand hit four times normal, supplies were emptied three times in under 24 hours, and twenty percent of hydrants in elevated Pacific Palisades areas ran dry.

A former department chief engineer said the pump and storage system "was designed for a fire that might consume several homes, not one that would consume hundreds." That is the right engineering framing, and it is not a defense. It describes a design basis that was known and never revisited.

Water quality followed. Pasadena Water and Power issued a water alert on January 8 for debris and elevated turbidity, and the Los Angeles Department of Water and Power issued boil water notices for ZIP codes 90272 and 90402 over potential benzene and volatile organic compounds.

West Altadena

West Altadena got evacuation orders at 3 a.m. on January 8, hours after East Altadena. All but one of the nineteen confirmed Eaton Fire deaths were people who lived west of Lake Avenue, a historically Black neighborhood where the fire destroyed nearly half of all Black households in Altadena. By 4 a.m., 52,314 residents were under orders and 46,847 under warnings.

On January 9, an erroneous Wireless Emergency Alert intended for the Kenneth Fire zone went to nearly ten million Los Angeles area residents countywide. A county supervisor confirmed the error, and the county suspended the alerting system on January 10 in favor of the Governor's Office of Emergency Services.

California Attorney General Rob Bonta opened a civil rights investigation on February 12, 2026, asking whether race, age, or disability discrimination affected Eaton Fire preparation and response in West Altadena, focused on the delayed evacuation notices. It is a civil rights investigation, not a criminal one.

The idle line

Southern California Edison's own words, in its annual report for fiscal 2025 filed February 18, 2026 and repeated verbatim in the first and second quarter 2026 filings:

A viable explanation is that a de-energized idle SCE transmission facility in the preliminary area of origin was associated with the ignition of the fire and SCE is not aware of evidence pointing to another possible source of ignition.

Southern California Edison, Form 10-K for fiscal 2025, filed February 18, 2026

The Los Angeles County Fire Department leads the origin and cause investigation and has identified a preliminary area of origin, while CAL FIRE's incident page still lists the cause as undetermined. The Safety and Enforcement Division is investigating, the Los Angeles County District Attorney is investigating potential criminal violations, and the Department of Justice sued Edison in September 2025.

Months before the fire, regulators faulted Edison on two counts: it had fallen behind on transmission line inspections in high risk areas, and it sampled wind speed every ten minutes where neighboring utilities sampled every thirty seconds. A ten minute sample can miss a gust entirely, and gusts move conductors.

quite low in today's climate

Robert McCullough, electric utility consultant, on the Los Angeles grid's 56 mile per hour wind design standard

Edison's June 10, 2026 rebuttal concedes that absent more evidence its equipment could have been associated with the ignition, calls the plaintiffs' motion for a ruling without a jury trial "wrong on the facts and the law," defends idle lines as serving electrification, backup capacity, and regional growth, and calls the media coverage "one-sided, sensationalist." On the July 30, 2026 earnings call the chief executive said "liability is probable. Given everything here. We have taken accountability."

The Lachman Fire and the reignition theory

The Lachman Fire ignited at 12:17 a.m. on January 1, 2025 on the Skull Rock trail ridgeline and reached eight acres. Forward progress was stopped after 3:30 a.m., and a 36 hour patrol followed. The Bureau of Alcohol, Tobacco, Firearms and Explosives determined that "the cause of the fire was determined to be the introduction of an open flame (likely a lighter) to a combustible material such as vegetation or paper." The reignition theory is that the Lachman Fire smoldered underground, undetected, for days, and came back above ground when high winds returned on January 7.

Jonathan Rinderknecht, 29, was arrested October 8, 2025 on three federal felony counts of arson. Prosecutors alleged he was alone at the ignition location with no fireworks nearby, and digital evidence in the complaint included a chatbot query: "Are you at fault if a fire is lift [sic] because of your cigarettes?" United States District Judge Anne Hwang declared a mistrial on June 26, 2026 after the jury deadlocked ten to two for acquittal. Retrial is set for October 19, 2026.

Litigation posture

The lead Eaton case is Gursey v. Southern California Edison Company, No. 25STCV00731, Los Angeles Superior Court, Judge Laura Seigle. As of June 30, 2026 there were more than 2,000 lawsuits with roughly 32,000 plaintiffs, and about fifty bellwether plaintiffs were selected in spring 2026. The first bellwether trial is set for January 25, 2027, pretrial filings due January 4, and Judge Seigle rejected Edison's request to push it to August 2027. Edison's securities filings repeat identically across three consecutive filings that "a bellwether jury trial in the Eaton Fire litigation has been set for January 2027."

Two rulings in late June 2026 shaped the defense. On June 26 the court dismissed Edison's claims against three local water districts over inadequate water supply. On June 30 it let Edison pursue Los Angeles County over overgrown brush in Eaton Canyon as a dangerous condition, while rejecting claims against the Sheriff's Department and the emergency management office over delayed alerts on first responder immunity grounds.

That would mean that all throughout California, there would be a duty to go clear the brush. That is impossible.

Hon. Laura Seigle, Los Angeles Superior Court, questioning Edison's theory that Los Angeles County had a duty to clear brush, Courthouse News Service, May 28, 2026

Edison settled with two subrogation insurers at about 55 cents on the dollar and recorded $1.6 billion in losses as of June 30, 2026, while saying in the same filings that it is "currently unable to reasonably estimate a range of losses" and that it is "probable that Edison International and SCE will incur additional material losses." The $1.6 billion is settlements to date, not a ceiling. Do not publish a total Eaton exposure figure, because Edison has declined to book one.

The Wildfire Fund administrator reported about $21 billion in claims paying capacity available for the Eaton Fire, and Edison will be reimbursed for eligible third party damage losses above $1.0 billion.

Palisades victims sued the City of Los Angeles and its Department of Water and Power, alleging downed power lines caused a secondary fire around 10:30 p.m. on January 7. The department says the lines were manually de-energized around 2:15 p.m. that day. The consolidation structure, judge, rulings, trial date, and status of inverse condemnation claims against a municipal utility are in the punch list.

The direct compensation program, assessed honestly

Edison announced a direct compensation program in July 2025 and launched it that fall, excluding any admission of liability. It covers property loss, smoke and ash damage, business interruption, personal injury, and wrongful death. The published amendment framework offers $20,000 per adult resident affected by smoke, $5,000 per child, and up to $1.5 million per individual for families of deceased victims. Participants waive claims against Edison.

What is good about it. Thirty five days to an offer beats any litigated process in the country, and money reached people in the first year, not the fifth. Offers have run from $15,000 for a tenant with non burn damage to $15.1 million for a claimant with multiple properties, a range showing individual evaluation rather than a grid. It is open to people who cannot find or afford counsel: 62 percent of claims were filed without an attorney.

What is oversold. Edison reported an 82 percent satisfaction rating from nearly 120 paid participants, out of more than 12,000 represented, surveyed by the party writing the checks. Roughly $360 million paid against more than $750 million offered means most of the offered money has not been accepted. An attorney at a claimant advocacy organization has said publicly that applicants will likely receive about half of what they are owed, and that residents focused on shelter are not weighing litigation value. That statement is a paraphrase in the source; confirm it before quoting.

What is advocacy dressed as information. Edison publishes a three point "myths" message: claims submissions will not be used against claimants in litigation, charitable donations do not reduce offers, and lawsuits do not guarantee higher settlements because claims from the 2017 and 2018 fires "remain pending." The first two are useful consumer information. The third is a pressure tactic aimed at the plaintiffs' bar, delivered by the defendant.

Edison's press releases also carry named claimant testimonials. One Altadena resident said, "I knew that I was coming back. It's my dirt, my ash, my community. My choice was to go with the claims program and I was very pleased with it." That is a real person saying a real thing, and a utility selected testimonial in a utility press release. Label it that way wherever it appears.

Chapter 13

The Destroyed Home

The destroyed home is the easiest damages case in the file and is still routinely undervalued, because counsel treat the insurance adjustment as the starting point. It is not. It is a contractual payment under a policy almost certainly written short.

The baseline rule, and the burden shifting inside it

Under California's civil jury instruction on damage to real property, a plaintiff proves either the reduction in value or the reasonable cost of repair, and where both are shown recovers the lesser. Fair market value is "the highest price for the property that a willing buyer would have paid to a willing seller." Chapters fourteen through seventeen build on that measure.

When the plaintiff proves diminution in value, the defendant bears the burden of proving that repair cost is lower. When the plaintiff proves repair cost, the defendant must show that it exceeds property value. Choose which measure to prove first with that allocation in mind.

The exception that changes the number: restoration for personal reasons

Heninger v. Dunn (1980) 101 Cal.App.3d 858 holds that restoration cost may exceed diminution in value where "there is a reason personal to the owner for restoring the original condition" or "where there is reason to believe that the plaintiff will, in fact, make the repairs." For destroyed shade or ornamental trees the owner may recover "the fair cost of restoring his land to a reasonable approximation of its former condition, without necessary limitation to the diminution in the market value of the land," if the cost is "not unreasonable in relation to the damage inflicted and the value of the land prior to the trespass."

The exception does not require a unique property, only a genuine desire to restore and a reasonable cost.

Kelly v. CB&I Constructors, Inc. (2009) 179 Cal.App.4th 442 applies it. A negligently sparked brush fire damaged a 34 acre ranch worth $1.6 to $1.8 million. The jury awarded $2,629,810 in restoration and repair, $375,000 in tree damage, $99,000 in lost rental income, and $543,000 in annoyance and discomfort, for $3,646,810, adjusted to $4,721,014.12 after doubling and fees. Restoration costs exceeding pre fire value by 67 percent were affirmed as reasonable given the owner's intent to return as a primary residence.

Personal property, and the proof problem nobody solves

Contents are valued at fair market value at the time of loss, typically below replacement cost, with an exception for items of peculiar value such as irreplaceable family documents. The California instructions are those on damage to, loss or destruction of, and loss of use of personal property.

Proof of contents loss is an inventory, the most abandoned task in wildfire recovery. Twenty four months after the Camp Fire, 22 percent of survivors were still working on it and 23 percent had decided not to submit one. Chapter seventeen gives the insurer requirement data.

A tort claim for contents does not require the insurer's form, but it does require proof, and a client who abandoned the insurance inventory has no starting document. Build it once, early, to serve both the claim and the case. Photographs, purchase records, credit card history, social media archives, and video walkthroughs made after the fact are all usable.

Loss of use, and why the rental figure is not the whole figure

Loss of use is measured by the reasonable rental cost of comparable property during reconstruction. Kelly included $99,000 in lost rental income. Two adjustments are routinely missed. The reconstruction period is the actual one, not the period the insurer allowed; twelve months after the January 2025 fires only 41 percent of intended rebuilders had applied for a permit. And the fire itself distorts the comparable rental market inside a burn zone upward, as chapter sixteen shows.

Betterment, and how to defeat it

Defendants argue a rebuilt property is worth more than the destroyed one and seek a deduction. Courts distinguish code required improvements from discretionary upgrades. Do not concede deductions for unavoidable code compliance. A house rebuilt to current wildland urban interface code is not an upgrade the owner chose. It is the only lawful way to replace what was taken.

Where the offsets come from

Chapter seventeen works that last line through.

Chapter 14

When the View Burns and the House Does Not

We have to live in the scar. We're all still really struggling. This isn't over for us.

Julie Lawson, homeowner, Altadena, PBS NewsHour and the Associated Press, January 5, 2026

A house survives. The insurance company closes the file. The neighborhood is gone. The hillside behind it is black, the sixty year old oaks stand dead, and the property is worth ten to sixteen percent less than the week before the fire. That loss is measurable, the defendant caused it, and in most cases nobody claims it.

The dataset that isolates exactly this population

One year after the January 2025 Los Angeles fires, Realtor.com analyzed Cotality and Quantarium valuation data and broke out homes inside the fire zones by damage status. The untouched rows are the ones that matter.

Total value erased across both communities: $8.3 billion. Destroyed homes sold as vacant lots traded at roughly half their most recent pre fire purchase prices.

Proximity and visibility are priced separately

Stetler, Venn and Calkin, 2010, Ecological Economics, is the most directly on point study available. It analyzed 17,693 home sales in northwest Montana from 1996 to 2007 against 256 wildfires.

Larger fires, at or above 405 hectares, with a view produced stronger negative effects, and the effect persisted beyond seven years, the study's maximum observation window. The authors' framing is a gift to a plaintiff: "When burned areas [are] out of sight, wildfire risk appears out of mind." That study is the backbone of a viewshed damage opinion: it separates proximity from visibility and prices each, and the ten kilometer threshold and seven year persistence are both usable.

Mueller, Loomis and González-Cabán, 2009, Journal of Real Estate Finance and Economics 38(2):155-172, on repeated fires in Southern California. The first fire produced a 10 percent decline in house prices; a second in the same small area produced 23 percent, a statistically significant difference. Repeat exposure compounds.

The study the defense will cite. Bengali, Nechio and Stewart, 2024, Federal Reserve Bank of San Francisco Economic Letter 2024-22, found about a 2 percent value boost for being farther from past fires in 2018 to 2021, and near zero for 2008 to 2017. It analyzed properties three or more miles from fire perimeters, excluding the close in band where Stetler found the large effects. It measures risk perception at distance, not amenity destruction next door.

What a view is worth, and the honest limit of the analogy

Benson, Hansen, Schwartz and Smersh, 1998, Journal of Real Estate Finance and Economics, analyzed 687 single family sales in Bellingham, Washington in 1993. A full ocean view added 58.8 percent, a superior partial ocean view 30.8 percent, a good partial view 29.4 percent, a poor partial view 8.2 percent. Lake frontage added 126.7 percent and a lake view without frontage 18.1 percent. Distance decay for a full ocean view ran 68.3 percent at one tenth of a mile, 55.6 percent at half a mile, 44.7 percent at one mile, and 30.6 percent at two miles.

The doctrinal trap: there is no right to a view

California recognizes no right to a view. A structure "cannot be complained of as a nuisance merely because it obstructs the view from neighboring property." Venuto v. Owens-Corning Fiberglas Corp. (1971) 22 Cal.App.3d 116 rejected a tenant's claim that pollution obscured mountain views, and Wolford v. Thomas (1987) 190 Cal.App.3d 347 is to the same effect. California rejected the English ancient lights doctrine, and view protection exists only by contract, easement, covenant, or zoning.

A destroyed viewshed is not a standalone nuisance. It is an element of diminution in value flowing from a tortiously caused fire. A burned tree works the same way: not its own tort, but a component of property damage. The bridge is the principle stated in Rony v. Costa (2012) and carried into the California jury instruction on trespass to timber, that in considering the diminished value of an injured tree the finder of fact may account for lost aesthetics and functionality. The same logic extends to the parcel.

Trees have their own statutes, and one of them was taken away in 2020

California Civil Code section 3346(a) sets the multiplier: "For wrongful injuries to timber, trees, or underwood upon the land of another, or removal thereof, the measure of damages is three times such sum as would compensate for the actual detriment, except that where the trespass was casual or involuntary" a lesser multiplier applies. Code of Civil Procedure section 733 makes anyone who "cuts down or carries off any wood or underwood, tree, or timber, or girdles or otherwise injures any tree or timber on the land of another person" liable "for treble the amount of damages."

Salazar v. Matejcek (2016) states the structure in three tiers. For willful and malicious trespass the court may impose treble damages and must impose double. For casual and involuntary trespass the court must impose double. For trespass under authority, actual damages only.

What survives is the multiplier for intentional tree injury, and it reaches further than most practitioners realize. In Fulle v. Kanani (2017) 7 Cal.App.5th 1305, a neighbor cut six mature trees on an Encino property to half height to open his own valley view, and the jury awarded $27,500 in tree damage, $20,000 in repair, and $30,000 in past noneconomic loss. The court held that "annoyance and discomfort damages resulting from tortious injuries to timber or trees are subject to the damage multiplier under sections 733 and 3346," because Civil Code section 3282 defines detriment as "a loss or harm suffered in person or property." The multiplier reaches "the full measure of compensable damages for tortious injury to property," and is "not limited to the value of the timber." Hassoldt v. Patrick Media Group (2000) bars doubling or trebling where punitive damages are awarded under section 3294.

Other states, and why Washington is the best forum in the country for this claim

Oregon. ORS 105.810 gives treble damages against anyone who willfully injures or severs produce, or "cuts down, girdles or otherwise injures or carries off any tree, timber or shrub" without authorization, plus reasonable costs of litigation including investigation costs and attorney fees. A contract logger safe harbor limits recovery to actual damages where the logger worked under a signed written contract with a reasonably believed owner, verified metes and bounds boundaries, kept deed copies for three years, and had no notice of an adverse claim. ORS 105.815 gives double damages for casual or involuntary trespass. Permanent injury takes the before and after value difference; temporary injury takes restoration cost where that is reasonable or less than the value loss; diminution controls if restoration is disproportionate.

Washington. RCW 64.12.030 covers anyone who cuts down, girdles, or otherwise injures or carries off a tree on another's land without lawful authority: "Any judgment for the plaintiff shall be for treble the amount of damages claimed or assessed." RCW 64.12.040 mitigates to actual damages for casual or involuntary trespass. The purpose is to stop people from removing another's merchantable trees on the gamble that the job pays if only actual damages follow.

Washington is the best jurisdiction in the country for this claim because Pendergrast v. Matichuk (2016) holds that emotional distress damages are recoverable under the timber trespass statute and are themselves trebled: $115,000 in emotional distress plus $8,510 economic, both trebled. Happy Bunch, LLC v. Grandview North, LLC (2007) treats boundary trees as common property with damages apportioned by the percentage of trunk on the plaintiff's land.

Colorado. No dedicated timber trespass multiplier statute was located. Board of County Commissioners of Weld County v. Slovek, 723 P.2d 1309 (Colo. 1986) controls: diminution in market value is the general rule, with trial court discretion to apply repair or restoration cost where that better reimburses the actual loss.

Texas. Gilbert Wheeler, Inc. v. Enbridge Pipelines (East Texas), L.P., 449 S.W.3d 474 (Tex. 2014) draws the line. Injury is permanent if it cannot be repaired, or if repair is possible but the harm is substantially certain to recur repeatedly, continually, and regularly; temporary injury is repairable and recurs only occasionally and irregularly. Permanent injury takes difference in value, temporary takes cost of repair, and an economic feasibility exception makes the injury permanent, capping recovery at fair market value reduction, when repair costs dramatically exceed the diminishment. The intrinsic value exception for ornamental trees is directly useful: Gilbert Wheeler broadened Porras v. Craig (1984) so that intrinsic value is recoverable not only where market diminution is zero but where there is "so little diminishment of that value that the loss is essentially nominal," compensating the tree's "ornamental (aesthetic) value and its utility (shade) value."

How an appraiser actually values a tree

Stumpage is the standing merchantable timber value, the correct measure for a commercial timber stand and a catastrophically wrong one for a sixty year old oak shading a house. Heninger is the authority for the shift: for shade and ornamental trees the owner recovers the fair cost of restoring the land to a reasonable approximation of its former condition.

The appraisal method is set out in the Council of Tree and Landscape Appraisers Guide for Plant Appraisal, tenth edition, which recognizes three approaches: cost, income, and sales comparison. For large specimen trees the trunk formula technique under the cost approach governs, extrapolating the cost to purchase the largest commonly available nursery plant up to the size of the appraised plant.

1. Compute cross sectional area of the trunk: radius squared times 3.14.
2. Compute basic value: cross sectional area times the unit cost per square inch.
3. Derive unit cost from the regional wholesale price for the largest commonly available transplanted tree. A published worked example uses Indiana's largest deciduous nursery stock at 3.5 inch caliper, yielding $37.67 per unit.
4. Depreciate basic value by ratings for condition, meaning health, structure, and form; functional limitations, meaning site and genetic factors; and external limitations, meaning environmental threats beyond the owner's control.

The same worked example produces a reproduction value of $3,400 for a 14 inch sugar maple in good condition. Multiply that across a mature residential canopy and the number stops being decorative.

Practitioners work with arborists to value a tree by type, size, age, condition, placement, function, and aesthetic contribution, with recovery reaching landscaping, statuary, wooden structures, and soil remediation.

Annoyance and discomfort is the head of damage that fits

The doctrine originates in Kornoff v. Kingsburg Cotton Oil Co. (1955) 45 Cal.2d 265, 272, permitting occupants to recover for annoyance and discomfort from nuisance or trespass. The California jury instruction permits recovery for "annoyance and discomfort, including emotional distress or mental anguish, caused by the injury to [his/her] peaceful enjoyment of the property." Compensable annoyance arises from "physical discomfort, irritation, or inconvenience caused by odors, pests, noise, and the like." The restrictions on emotional distress damages that apply in breach of contract or negligence cases do not apply to property tort invasions.

That is the one head of damage that pays for the lived experience of a surviving house: the smell, the dust, the emptied street, the three years of jackhammers, the neighbors who did not come back. It does not require proof that the market noticed.

Who can recover. Kelly v. CB&I Constructors holds that "a nonresident property owner who merely stores personal property on the premises is not entitled to recover annoyance and discomfort damages," and requires "immediate and personal possession, as a resident or commercial tenant would have." Hensley v. San Diego Gas & Electric Co. (2017) 7 Cal.App.5th 1337, from the 2007 Witch Creek fire, holds that homeowners may recover "even though [they were] not physically present to see the fire ravage [their] house." Possession is required; presence during the damage is not.

Emotional distress is folded in, not separate. Under Hensley, emotional distress damages are not distinct from annoyance and discomfort damages. Plaintiffs therefore need not satisfy the independent duty requirement or the bystander requirements that govern negligent infliction of emotional distress. The distress rides on the property tort.

Stigma, and the element that a burn scar uniquely satisfies

Stigma has two definitions and they get mixed up constantly. The useful one is residual: a persisting loss in market value of a property that was damaged and then repaired. The loose one treats stigma as a synonym for ordinary diminution, which is usually what a court means when it says it awarded stigma damages.

The majority rule denies recovery for residual stigma. Per Mehlenbacher v. Akzo Nobel Salt, Inc., 71 F. Supp. 2d 179 (W.D.N.Y. 1999), causing value to diminish is not itself a basis for tort liability, and most courts deny recovery absent physical damage or where repair is practicable.

The gateway is the three part test from In re Paoli Railroad Yard PCB Litigation, 35 F.3d 717 (3d Cir. 1994): diminution claims are permitted where "(1) defendants have caused some (temporary) physical damage to plaintiffs' property; (2) plaintiffs demonstrate that repair will not restore the value to its prior level; and (3) plaintiffs show that there is some ongoing risk to their land."

Element three is the hinge, and a burn scar satisfies it better than almost anything else. Post fire debris flow and flood risk is a documented, quantified, multi year physical hazard, not a speculative fear, and chapter sixteen supplies the mechanism and numbers. That is a better record than most contamination stigma plaintiffs can build.

Two courts allowed it. Walker Drug Co. v. La Sal Oil Co., 972 P.2d 1238 (Utah 1998): excluding stigma testimony was error, and an expert's inability to state the amount precisely is not grounds for exclusion. Bonnette v. Conoco, Inc., 837 So. 2d 1219 (La. 2003): diminution affirmed despite completed remediation. Others restricted it. Orkin Exterminating Co. v. DelGuidice, 790 So. 2d 1158 (Fla. Dist. Ct. App. 2001) allows it only where repair is impracticable and substantial economic waste. Hammond v. City of Warner Robins, 482 S.E.2d 422, and Houston Unlimited, Inc. Metal Processing v. Mel Acres Ranch, 443 S.W.3d 820 (Tex. 2014) called it too remote and speculative. Exxon Mobil Corp. v. Albright, 433 Md. 303 (2013) rejected it on double recovery grounds. In California, Santa Fe Partnership v. ARCO Products Co. (1996) 46 Cal.App.4th 967 limits damages for contaminated property and is the leading obstacle. Read it before relying on this theory in California.

Four mechanisms, and why you separate them

The American Bar Association names four pathways by which environmental harm depresses value. Keeping them separate is the difference between an admissible opinion and an excluded one.

1. Physical damage. One time destruction.
2. Avoidance costs. Ongoing mitigation with operating and maintenance costs.
3. Loss of use and enjoyment. Conditions that cannot be controlled or eliminated.
4. Stigma and risk uncertainty. The subjective potential of a purchaser to reduce willingness to purchase or to discount the property due to negative perception rather than actual contamination.

The econometric methods are the event study, comparing prices before and after disclosure, and difference in differences, comparing the affected area to a similar unaffected benchmark. The standing appraisal text is Bell, Anderson and Sanders, Real Estate Damages: Applied Economics and Detrimental Conditions, Appraisal Institute.

How to actually prove it, step by step

Step one: refuse the median. Post fire medians are contaminated by vacant lot sales. Altadena's median sale price peaked at $1.5 million in September 2023, fell 43 percent to $710,000 by March 2025, and recovered to $990,000 by February 2026, while days on market went from seven to fourteen before the fire to just over two months. One year out, the Palisades median fell from $3.7 million to $1.9 million and Altadena's from $1.35 million to $650,000. A Realtor.com economist cautions that vacant lot sales drag those medians down and are "in line with the typical listing price for lots." They do not measure surviving home value, and investors buying 30 to 40 percent of vacant lots distort the comparable set further. A diminution opinion built on a burn zone median will not survive cross examination.

Step two: build a paired sales or hedonic analysis that segregates improved parcels. Compare surviving improved parcels inside the burn footprint against matched improved parcels in an unaffected benchmark area with similar pre fire price levels, age, size, and school attendance. Difference in differences is the standard method and the one a court will recognize.

Step three: separate the four mechanisms and price each. Physical damage, meaning smoke and ash, goes in chapter fifteen's bucket. Avoidance costs are air filtration, soil testing, and repeated cleaning. Loss of use and enjoyment is the annoyance and discomfort claim. Stigma is the residual after repair. Overlap is double recovery, and it is the first thing the defense will find.

Step four: quantify the setting, not the view. Do not price the panorama. Price what is measurable: the tree canopy lost on the parcel, appraised by trunk formula; the tree canopy lost in the immediate viewshed, established by pre fire and post fire imagery; the proximity to the burn perimeter, priced against Stetler; the persistence window, set at seven years or more from the same study; and the documented debris flow hazard, which is both an avoidance cost and a Paoli element three.

Step five: prove the ongoing risk with a public document. The United States Geological Survey publishes post fire debris flow hazard assessments for major burn scars, and county flood control districts publish inundation mapping. Neither is plaintiff generated, and they establish element three without a retained expert.

Step six: plead annoyance and discomfort for every occupant, not only every owner. The requirement is immediate and personal possession. A tenant qualifies, a commercial tenant qualifies, and an owner who was out of town during the fire qualifies under Hensley. An out of state owner storing furniture does not, under Kelly.

They cannot ignore houses that are burned down to the ground. They can't ignore those. So the ones that they try to save money on are the neighbor whose house looks fine, but is not fine.

Dylan Schaffer, plaintiffs' attorney, Kerley Schaffer LLP, ABC7 Los Angeles, June 29, 2026

When I am cutting my grass, I have my head down.

Daniel Quiroz, sanitation worker for the City of Los Angeles and homeowner on Olive Avenue in Altadena, whose house survived, The Christian Science Monitor, June 20, 2025

No adjuster's estimate has a line for a man who will not look up from his own lawn. The annoyance and discomfort instruction covers it exactly.

Chapter 15

Smoke, Ash, and Char

No accredited United States standard governs the assessment of a wildfire smoke damaged home. The only accredited fire and smoke restoration standard expressly excludes wildfire, and the industry document written to fill the gap makes objective clearance testing optional.

The regulatory gap, stated exactly

The Standard for Professional Fire and Smoke Damage Restoration, S700, was accredited by the American National Standards Institute for the Institute of Inspection, Cleaning and Restoration Certification and published January 3, 2025, four days before the Eaton and Palisades fires ignited. Its scope reads: "The fire event can occur within the building, an adjoining building(s), or building(s) in the vicinity impacted by an external or internal fire, other than wildfires."

The substitute is the joint Technical Guide for Wildfire Restoration, Version 2, December 9, 2025, published by that certification institute with the Restoration Industry Association and the Cleaning Industry Research Institute. It is a guide, not an accredited standard, and its clearance provisions are the problem. Restoration Completion Evaluation rests on "visual and olfactory inspection" confirming the property is "free of visible residue and smoke odors." Post-Restoration Verification, meaning third party testing, is optional and "reserved for specific circumstances."

What the contaminants actually are

The Technical Guide defines five categories of combustion byproducts of concern.

• Soot, char, and ash. The primary particulates.
• Volatile organic compounds. Gaseous combustion byproducts that can be forced into the building envelope.
• Semi volatile organic compounds. Compounds that adsorb or condense onto particulates.
• Heavy metals. Copper, lead, barium, zinc, and arsenic.
• Asbestos. In pre-1970 structures with burn damage.

What burns is why. A survivor coalition's inventory: termite treated wood into arsenic; wool rugs, nylon carpets, paper, and upholstered furniture into cyanide; electric car batteries into lithium; old microwaves into beryllium. Add magnetite, pesticides, and plastics. A structure fire burns one building's contents. A wildland urban interface fire burns a neighborhood's contents and deposits the residue inside the houses that did not burn.

a wildland-urban interface fire doesn't produce ordinary smoke. We know the toxins are there. We don't yet have binding rules.

Michelle Meyers, partner, Singleton Schreiber, attorney representing Eaton Fire homeowners, CalMatters, May 5, 2026

The measured levels

That figure is the empirical basis for two claims: residual stigma, and a claim that the remediation the insurer paid for did not work.

Why ordinary cleaning fails

Particle size. Wildfire smoke is dominated by fine particles at or below 2.5 micrometers in aerodynamic diameter, with a substantial ultrafine fraction below 0.1 micrometers. Particles that small penetrate wall cavities, mechanical systems, and the pore structure of finish materials. They are not on the surface.

Reservoir and re-emission behavior. Volatile organic compounds from wildfire smoke stay elevated for days and weeks, repeatedly partitioning between indoor air and surfaces. The evidence review found surfaces became a net source of acrolein emissions at 4.8 hours, with compounds persisting for weeks to months. Portable air cleaners cut volatile organic compounds by 3.8 to 5 percent, and the gain was gone within thirty minutes. Ventilation achieved 55 to 85 percent reductions, and concentrations rebounded close to initial levels within hours.

Improper cleaning makes it worse. The Technical Guide says cleaning done wrong "can cause further spread and sorption of combustion byproducts." Dry wiping a soot film drives it into the substrate.

Some things work. In the same review, combined vacuum, mop, and dust cut airborne formic acid by 50 percent and formaldehyde by 32 percent, wet wiping cut polycyclic aromatic hydrocarbons on glass by 62 to 71 percent, and machine laundering removed 48 percent of those compounds from cotton. Evidence gaps remain on settled dust and trace metals.

Materials in our homes act like a sponge. They soak up a lot of these VOCs into painted surfaces.

Joost de Gouw, chemist, University of Colorado Boulder and the Cooperative Institute for Research in Environmental Sciences, Boulder Reporting Lab, December 6, 2022

Testing methods, and the one your client's adjuster will use

The Technical Guide recognizes five methods: tape lifts for particle morphology, wipe sampling for metals and semi volatile organic compounds, micro-vacuuming for porous surfaces, bulk sampling for deep penetration, and air sampling by thermal desorption gas chromatography mass spectrometry. Analysis targets char, ash, and soot morphology, organic markers unique to biomass combustion, and metal concentration ratios.

Survivors' criticism of insurer testing is technically correct. Adjusters have been documented "using a piece of packing tape to make a 'tape lift' and sample just a small amount," where "testing standard for lead is based on the amount of lead that is in a 1 x 1 foot frame, which is required to be collected by wet swabs wiped multiple times." A tape lift is a qualitative morphology tool, not a quantitative metals measurement, and using it as one produces a false negative by design.

Name these other standards in the report: the National Air Duct Cleaners Association assessment, cleaning and restoration standard for heating, ventilation, and air conditioning systems; the Environmental Protection Agency lead dust hazard standards under Toxic Substances Control Act sections 402 and 403; the Occupational Safety and Health Administration lead in construction standard at 29 Code of Federal Regulations 1926.62; and the asbestos standard at 1926.1101.

The protocol to demand

1. Pre cleaning inspection covering interior and exterior, basements and crawl spaces, heating and cooling ducts, and wall cavities.
2. A written remediation plan addressing every finding in that inspection.
3. Post cleaning verification by an independent third party.
4. Repeat rounds as needed until verification passes.

Testers should be licensed industrial hygienists or indoor air quality professionals with urban wildfire experience, independent of the remediation contractor. A contractor who both performs and certifies the work has an unmanaged conflict.

The denial pattern, documented

The survivors wrote the California Smoke Claims and Remediation Task Force on February 4, 2026. Insurers tell homeowners their homes need only "a light clean," send adjusters who "walk through, sniff the air, and 'wipe it down. Run an air filter,'" and "delayed claims, then unexpectedly cut off standing home survivors'" additional living expense and fair rental value payments.

A United Policyholders webinar of December 18, 2025 documented four patterns: declining comprehensive pre remediation testing, including air, volatile organic compound, and in wall cavity sampling; third party vendors checking only basic equipment function rather than contaminant intrusion; calling contents cleanable despite a hygienist's disposal recommendation; and terminating additional living expense payments to push residents back into unremediated homes.

Twelve to thirteen months after the January 2025 fires, in the standing home and partial loss cohort, 74 percent had not moved back. Fifty-two percent had an insurer threaten to cut temporary rent payments with the claim still open. Eighty-eight percent had their homes tested before remediation and 46 percent paid out of pocket for it, with out of pocket spending at 32 percent for restoration work, 25 percent for cleaning and remediation, and 21 percent for re-testing.

We are trying everything that we can to avoid litigation. We shouldn't have to sue our insurance company.

Elissa Ashwood, homeowner, Lachman Lane, Pacific Palisades, ABC7 Los Angeles, June 29, 2026. Her claims examiner deemed the home habitable despite an independent assessment finding significant thermal, non thermal, and chemical contamination

The coverage decision that changed the law

Aliff v. California FAIR Plan Association, Case No. 21STCV20095, Los Angeles Superior Court, Judge Stuart M. Rice, decided June 25, 2025, held that the FAIR Plan's policy violates California Insurance Code section 2071 by providing less coverage than the state mandated standard form fire insurance policy. It struck down the plan's post-2017 requirement of "permanent physical changes" to property and its sight and smell standard for smoke claims, along with the dispute resolution procedures tied to them.

The plan's own 2017 notice had said the new definition "will result in the denial of claims that might have been paid under prior policy wording." Judge Rice answered: "This notice seems to admit that the CFP Policy is less favorable to insureds than the Standard Form Policy."

The operative rule, aligning with the California Supreme Court's 2024 decision in Another Planet Entertainment LLC v. Vigilant Insurance Co., is that direct physical damage need not be visible to the naked eye.

The baseline coverage rule that follows: wildfire debris damage, meaning smoke, ash, char, and particulates, is fire damage, which must be covered under Insurance Code section 2071.

Regulatory enforcement, and the legislation it produced

California Department of Insurance Bulletin 2025-7, issued March 7, 2025, ordered insurers to fully investigate smoke damage claims. They "cannot summarily deny smoke damage claims without a thorough investigation," evidence of smoke damage "must be fully and fairly investigated," and where professional testing is warranted, "we expect the insurance company to contract and pay for these services." It is unreasonable "for the insurer to require the insured to incur substantial costs to investigate their own claim."

On July 31, 2025 the Department brought an Order to Show Cause against the FAIR Plan under Insurance Code section 790.03, alleging misrepresentation of policy terms, failure to investigate fairly, and denial of legitimate claims without reasonable basis, and identifying at least 418 violations. A March 2021 market conduct exam had already reviewed 259 claims and found 118 violations tied to smoke damage denials, documenting the pattern four years before the Los Angeles fires.

On May 4, 2026 the Department took legal action against State Farm. A market conduct examination of 220 claims found 398 violations across 114 claims, 51.8 percent of the sample, and consumer complaints added 34 more. State Farm policyholders filed about 11,300 residential claims from the Los Angeles fires, nearly one third of the 38,835 total, and smoke claims made up nearly half of all consumer complaints to the Department. The violations run through the whole claims process: failure to begin investigation within 15 days, accept or deny within 40 days, or pay within 30 days; unreasonably low offers; repeated adjuster reassignment, which the Department called "adjuster roulette"; failure to provide required written denials for testing; and misrepresentation of policy provisions. Penalties run to $5,000 per violation and $10,000 for willful violations. The Department called it the largest amount pursued this century after a wildfire disaster.

Wildfire survivors came to us for help, and we followed the facts. Our investigation found that State Farm delayed, underpaid, and buried policyholders in red tape at the worst moment of their lives.

Ricardo Lara, California Insurance Commissioner, California Department of Insurance press release, May 4, 2026

The California Smoke Damage Recovery Act

Assembly Bill 1795, the Smoke Damage Recovery Act, was unveiled February 11, 2026 by Commissioner Lara and Assemblymember Mike Gipson and advanced March 13, 2026 as the nation's first smoke damage standards bill. It would set California's first enforceable standards for smoke damaged homes: science based, health driven standards for inspection, testing, and restoration; uniform claims handling; required restoration protocols; health based occupancy guidelines; designated enforcement agencies; and an early action provision letting survivors use local public health agency standards while statewide standards are developed.

A companion bill, Assembly Bill 1642, would establish health clearance standards. It passed its first committee five to nothing, and the insurance industry formally opposes it. Assembly Bill 1795 references industry standards with acknowledged limitations, and the industry has said nothing about it.

The Department's Smoke Claims and Remediation Task Force reported in March 2026, acknowledging the absence of residential health standards for most contaminants found. Its composition drew criticism: two insurance industry lobbyists and three remediation professionals with insurer ties, and no fire survivors or independent toxicologists. More than 13,000 smoke damage claims were filed after the January 2025 fires.

Colorado reached the same conclusion first

Partners Environmental Consulting Inc. filed a draft report with the Colorado Division of Insurance on November 18, 2025 recommending standardized sampling protocols, enforceable residential cleanup standards, repair versus replace criteria, and reoccupancy clearance thresholds. It urged the state to "create a minimum standard for all houses impacted by wildfires to establish a clear yes/no decision process." Without one, "the result has been widely varying claims and an inefficient and sometimes indefinite cleanup process."

Two University of Colorado studies of smoke damaged but structurally intact Marshall Fire homes found volatile organic compounds falling after the fire and then lingering far longer than expected, with elevated benzene, copper, zinc, arsenic, and industrial pollutants. A health survey of 859 residents found over half of standing home survivors reporting health problems from poor indoor air quality. A separate unified research survey in August 2022 covered 831 respondents living away from smoke damaged homes, and more than 40 percent reported itchy or watery eyes, sore throat, headache, dry cough, and a metallic taste.

An anxiety-induced physical symptom is still a physical symptom, even if others aren't experiencing it.

Courtney Welton-Mitchell, social psychologist, Colorado School of Public Health, Boulder Reporting Lab, December 6, 2022

Chapter 16

The Downstream Years

A wildfire case does not end when the fire is out. The burn scar changes watershed hydrology for years. Agricultural losses surface at harvest and again at fermentation. Businesses fail on a lag. Evacuation costs run long after the smoke clears. These damages arrive after the complaint is drafted, which is why they are so often left out of it.

Why a burn scar floods

The United States Geological Survey states the mechanism plainly: "After a fire, vegetation is removed and soil properties change, reducing the ground's ability to absorb rainfall." Post fire debris flows are "fast-moving mixtures of water, mud, rocks, and vegetation that can surge downslope like flowing concrete," and they are "typically triggered by short bursts of intense rainfall." Even modest rainstorms can set off dangerous flash floods and debris flows in steep burned areas.

Montecito, and the three week interval that forecloses a superseding cause defense

Kean and colleagues documented the January 9, 2018 Montecito debris flow in Geosphere.

A defendant arguing the rainstorm was an independent, superseding cause has to explain a twenty one day gap between its fire and the flow, in catchments its fire denuded, in the ordinary rainy season. Southern California Edison settled Thomas Fire and debris flow lawsuits: one reported payment of $360 million covered wildfires and the mudslide, and a settlement over a billion dollars was reported in September 2020. Verify the allocation between fire and debris flow claims before citing either figure.

The multi year window

Elevated debris flow hazard does not end with the first rainy season. The literature on how the hazard changes as burned areas recover supports two things: the Paoli element three ongoing risk requirement from chapter fourteen, and an avoidance cost claim for the years of mitigation a homeowner must fund. The specific duration is in the punch list. Get it from the primary source before putting a number in a report.

The rest you can establish today from public documents. The United States Geological Survey publishes post fire debris flow hazard assessments for major burn scars, and county flood control districts publish inundation mapping. Those show an ongoing, agency recognized hazard on the parcel, which is what the doctrine requires.

Business interruption

Lost profits are governed in California by the jury instruction on lost profits, and the proof burden is heavier than for property damage. Three problems recur.

The baseline. A business inside a burn zone often had a declining or seasonal revenue trend before the fire, and the defense will build its counterfactual on the worst available window. Get three years of tax returns and monthly revenue before selecting the baseline period.

The recovery curve. A restaurant in a neighborhood where 9,400 structures burned does not return to pre fire revenue when it reopens. It returns to the revenue supportable by the population that came back, which chapter eleven shows can be sixteen percent short four years later.

Mitigation. Defendants argue the business could have relocated. The answer is usually that the customer base was geographic and did not move with it, but that has to be proved rather than asserted.

Vineyard smoke taint, the best quantified agricultural loss

The chemistry is why this loss is provable. Smoke deposits volatile phenols, principally guaiacol, 4-methylguaiacol, 4-ethylguaiacol, and eugenol, through the grape's waxy cuticle. Those compounds glycosylate, bonding with grape sugars into non volatile forms that are undetectable in the fruit. During fermentation the glycoside bonds break and the volatile phenols become perceptible again, producing a smoky, ashy finish. Red wines are more vulnerable, because harvest coincides with peak fire season and skin contact during fermentation increases extraction. The loss is latent at harvest and shows up months later in the tank, so a grower who tested clean at the crush pad can still lose the vintage.

The two totals, $601 million statewide and $535 million across three counties, come from different scopes and methods. Present both. Do not average them.

Three structural problems make this a tort claim rather than an insurance claim. Testing cost $160 to $360 per sample, and laboratories ran four to five weeks behind in 2020, past optimal harvest. No federal crop insurance threshold exists for guaiacol or 4-methylguaiacol. And roughly 95 percent of California wine grapes sell under contract, while most 2020 contracts had no smoke exposure language, leaving growers and wineries to litigate rejection.

There are so many variables. Freshness of the smoke, number of times exposed, variety of grape, the list goes on. There's so much we don't know.

Anita Oberholster, enology specialist, University of California Davis, quoted in the Smoke Exposure Research Act of 2023 one-pager

The federal response authorized $32.5 million over five years for smoke taint research, and Oregon State University received a $7.65 million grant to study smoke impact on grapes and wine. Science is moving faster than insurance standards, and the widening gap between provable agronomic loss and covered loss is a tort claim.

Livestock

The federal programs are the Livestock Indemnity Program and the Emergency Livestock Relief Program, run by the Farm Service Agency. On May 29, 2025 the Department of Agriculture committed $1 billion to livestock producers hit by drought or wildfire in 2023 and 2024, and on September 12, 2025 another $1 billion for flood and wildfire impacted producers.

The research for this book found no case law on livestock valuation in wildfire tort litigation and no aggregate California livestock loss figures for the 2025 fires. This section is thin and is in the punch list. In a livestock case the valuation framework has to come from agricultural appraisal practice rather than wildfire precedent.

Evacuation and loss of use

Evacuation costs are documented and routinely left out of a demand. In the survivors' letter to the California task force, standing home survivors reported moving an average of six times a year at a 30 to 80 percent premium over regular rentals, with contents removal running $10,000 to $100,000 or more. Those are receipted, out of pocket numbers, and they belong in the loss of use claim alongside the rental differential.

For a total loss household, loss of use runs for the actual reconstruction period. For a standing home household, it runs until the house is verifiably habitable. Chapter fifteen shows how long that fight can take.

Chapter 17

The Insurance Gap

"I have insurance, why worry" is the most expensive sentence a homeowner says. Seven out of ten who say it are wrong, and the average shortfall runs about twenty percent of the cost to rebuild. The error is not theirs. Software made it, in fifteen minutes, at the point of sale, years earlier.

The 2025 Los Angeles numbers, at twelve months

United Policyholders surveyed 675 households and 1,758 survivors between January 6 and February 17, 2026: 62 percent Eaton, 38 percent Palisades, 97 percent homeowners, 68 percent total loss.

At six months the same fires showed 62 percent of total loss respondents underinsured, 33 percent unsure, 5 percent with sufficient coverage. Underinsurance rose from 62 to 69 percent between six and twelve months as people got real construction bids. It is not discovered at the fire. It is discovered nine months later, at the contractor's table.

Marshall, and nineteen years of the same answer

United Policyholders' twelve month survey of 574 Marshall households found 80 percent with not enough insurance and an average shortfall of $110 per square foot. Its twenty four month survey of 451 households found 79 percent still insufficient, 15 percent adequate, and 31 percent unsettled on the dwelling portion at two years.

Nineteen years, a dozen fires, two states, same answer. The Camp Fire shows the same drift: the shortfall grew from $163,000 at twelve months to $203,000 at twenty four, for the same reason it grew in Los Angeles.

The census level evidence

The strongest citation is Kenneth S. Klein's 2026 article in the Lewis & Clark Law Review, which rests on regulator data rather than self report: California Department of Insurance aggregated fire claims data from 2018 to 2023, obtained through four Public Records Act requests, covering at least 98 percent of all fire insurance claims in California and more than 74,000 fire damaged or destroyed owner occupied homes. That is a census, not a sample.

Slightly more than seven in ten homeowners are likely underinsured, by an average of roughly twenty percent. Homes destroyed in wildfire are underinsured roughly eight times in ten, averaging a twenty five percent shortfall. Homes with an ordinary single house fire are underinsured 68 percent of the time, averaging seventeen percent.

The mechanism: fifteen minutes of software

The dominant factor explaining underinsurance appears to be the inaccuracy of the reconstruction estimate the insurer provides at point-of-sale.

Kenneth S. Klein, Louis and Hermione Brown Professor of Law, California Western School of Law, Lewis & Clark Law Review, Vol. 30.1, 2026

Insurers set the dwelling limit, Coverage A, using algorithms from two vendors. The consumer answers a pre-fill questionnaire of twelve to twenty four questions, some objective, such as square footage, age, and address, and some judgment calls, such as kitchen quality and construction grade. The quote usually takes fifteen minutes or less, and the resulting point of sale reconstruction estimate is low between 75 and 95 percent of the time, with the average underestimate above thirty percent.

The product changed and the customer's mental model did not

Guaranteed replacement cost was the pre-1990s norm: a destroyed home with that coverage got rebuilt no matter the cost. Between 1988 and 1997 almost all insurers moved to replacement cost value, which caps what the policy pays toward reconstruction, and that has been the standard since roughly 1996. The Federal Housing Finance Agency requires replacement cost value, not actual cash value, for compliant mortgages. Actual cash value insures only the depreciated value of a structure.

Homeowners' mental model of insurance was formed under guaranteed replacement cost and never updated. They think they bought "my house gets rebuilt." They bought "a number gets paid."

The four mechanisms that turn a low limit into a catastrophe

Extended replacement cost endorsements do not fix it, because they are a percentage of the wrong number. The endorsement expands available Coverage A by a stated percentage, and some policies include it automatically if the property is insured at 100 percent of the point of sale estimate. Colorado mandates a minimum of 50 percent of Coverage A. California requires all homeowner policies to disclose demand surge and describe the endorsement as the protection against it. Klein's Marshall Fire finding settles it: the construction cost spike "was largely absorbed by extended replacement cost provisions. Hence, policyholders were underinsured on average, not because of cost inflation, but because they purchased an insufficient amount of coverage A."

Ordinance and law limits are set far below observed need. The statutory minimum for building code upgrade coverage is 10 percent of Coverage A in California and 20 percent in Colorado. The actual purchased average, across 20,590 replacement cost policies from 2021 to 2023 with fire claims, is 38.83 percent. California's legal minimum is roughly one quarter of what the market buys.

Demand surge is not temporary. One major vendor reports that "almost regardless of the year, location, or type of catastrophe event, demand surge expresses as a cost increase of 15 to 30 percent," lasting six to twelve months; another estimates 20 to 30 percent. Observed values were 22 percent after Marshall and 16.7 percent after the North Bay fires. Klein: "While a classical definition of demand surge describes it as temporary, there is little to no evidence in the market that once residential construction prices surge, they ever meaningfully recede."

Additional living expense runs out before the rebuild finishes.

Twelve months after the fire, fewer than half of would be rebuilders had a permit application in. An application is not a permit, a permit is not a foundation, and a foundation is not a certificate of occupancy. Against that, a third of Marshall extension requests were denied at the moment the rent money stopped.

The 46 percent figure is a clean, citable example of policy terms operating against the law with nobody catching it. Nearly half of surveyed Marshall households reported dollar limits on temporary housing, and a Colorado statute prohibits them.

Contents: the inventory that defeats people

Twelve months after the January 2025 fires, 53 percent of insurers required an item by item listing of everything destroyed, 24 percent required purchase documentation, 15 percent required the insurer's own form, 40 percent allowed grouping of similar items, 14 percent imposed a submission deadline, 18 percent sent a contents specialist, 16 percent offered a cash out settlement, and only 23 percent explained the depreciation process. At six months, 57 percent had reported the item by item requirement.

Only 27 percent received 100 percent of contents benefits without an inventory requirement. In the 2020 California wildfires only 11 percent received full contents benefits without itemization, and in the 2020 Colorado wildfires 55 percent never received contents benefits at all without submitting an inventory. At twenty four months after the Camp Fire, 32 percent got a full waiver; of the rest, 55 percent submitted an inventory, 22 percent were still working on it two years later, and 23 percent were never going to submit one.

The personal property inventory is a total nightmare. I spent months and months on it, only to find it wasn't totaling what insurance already paid.

D.T., Eaton Fire survivor, United Policyholders 2025 Los Angeles Wildfires 12-Month Survey, March 2026

Depreciation holdback, the quiet one

Contents are paid at actual cash value first, and recoverable depreciation is released only after the policyholder buys the replacement and proves it. A survivor who cannot front the cash never sees the second check. Thirty percent of Marshall survivors said their insurer never explained depreciation or holdbacks, and only 23 percent of Los Angeles survivors had it explained at all.

The mortgage trap

The best hard evidence is a Federal Reserve Bank of Philadelphia working paper, "California Wildfires, Property Damage, and Mortgage Repayment," by Biswas, Hossain, and Zink, working paper 23-05, published March 2023 and revised November 2023. It merges property level damage inspections with regulatory mortgage data from 24 large banks, across 79 California wildfires from 2013 to 2020 and 81,940 mortgage observations. It uses actual property level damage inspections rather than fire perimeters, and undamaged properties inside fire perimeters showed no significant change in delinquency or prepayment.

Two thirds of burned out homeowners pay off their mortgages within two years, and most are not selling. The insurance check goes to the lender first, the loan is extinguished, and the owner keeps what is left of a settlement running 28 percent short of the rebuild. They lose the debt and the house together, and that debt was the cheap long term money that made the house affordable. They will not get it back at the same rate, or maybe at all, because carriers are non renewing in that market.

The availability crisis behind all of it

California's insurer of last resort has grown into something nobody designed. Official statistics through June 2026 show total exposure of $768 billion, up 11 percent from September 2025 and 250 percent from September 2022, and 696,562 policies in force, up 8 and 157 percent over the same periods. Written premium was $2.04 billion. New business in the first nine months of fiscal 2026 ran 151,061 policies, averaging 16,784 a month, down 25 percent year over year and still 208 percent above the September 2022 baseline.

On February 11, 2025 the plan levied a $1 billion assessment on member insurers to keep paying Los Angeles wildfire claims. Member companies bear half the cost under a prior agreement and may issue a temporary supplemental fee, but cannot fold assessment costs into future rates without the Commissioner's prior approval.

Wildfire survivors can't cash 'what ifs' to pay for food and rent, but they can cash FAIR Plan checks.

Ricardo Lara, California Insurance Commissioner, on the $1 billion FAIR Plan assessment, February 11, 2025

The Treasury Department's Federal Insurance Office analyzed an average of 49.3 million policies a year from 2018 to 2022, about 80 percent of private homeowners premium nationwide.

The national non renewal rate rose from 1.05 percent in 2018 to 1.20 percent in 2022. Premiums rose 8.7 percent above inflation nationwide, at least 14.7 percent in the top risk quintile, and fell 1.4 percent relative to inflation in the bottom. The Consumer Federation of America reported in 2024 that 7.4 percent of all homeowners in the country are uninsured, representing $1.6 trillion in unprotected market value.

California's non renewal moratoria have covered more than 600,000 consumers across 29 counties. A moratorium guarantees one renewal. It does not guarantee a second.

What to tell a client, before the fire

1. Get an independent reconstruction cost estimate from a licensed general contractor. Do not accept the carrier's figure.
2. Buy ordinance and law coverage at the observed market level near forty percent of Coverage A, not the statutory minimum of ten or twenty percent.
3. Buy an extended replacement cost endorsement, and understand it is a percentage of Coverage A, so it only helps if Coverage A is right.
4. Confirm additional living expense limits in both dollars and months, and confirm whether extensions are discretionary.
5. Build the home inventory now, with video and receipts, and store it off site.
6. Ask specifically how depreciation holdback works and what proof releases it.
7. Re-run the reconstruction estimate every two years, and after any renovation.

Chapter 18

The Human Cost

The mental health curve and the insurance curve bend at the same point. Depression falls through the first six months and rises again at eighteen, which is when additional living expense runs out, the underinsurance gap becomes undeniable, and the rebuild has stalled or been abandoned.

What the peer reviewed literature actually reports

The best single review is To, Eboreime, and Agyapong, "The Impact of Wildfires on Mental Health: A Scoping Review," Behavioral Sciences, Vol. 11, Issue 9, September 21, 2021.

Read the depression series in order

Three months: 25.5 to 33 percent. Six months: 10.4 to 17.1 percent. Eighteen months: 18.3 to 24.8 percent. Anxiety and post traumatic stress show the same eighteen month persistence. That is not a trauma response resolving. It is a second injury arriving on a delay, at the point the money runs out.

Substance use tracks the same shape. Adults: 17 percent alcohol use disorder or heavy drinking at three months, 22 percent at three years, 18.6 percent at ten years. College students eighteen months after the Fort McMurray fire: 15.5 percent high risk drinking, 13 percent problematic drug use, 4.4 percent moderate to high nicotine dependence. Those with likely generalized anxiety disorder were three times likelier to present with a drug related problem.

The controlled study on children

Brown and colleagues, "After the Fort McMurray wildfire there are significant increases in mental health symptoms in grade 7-12 students compared to controls," BMC Psychiatry 19:18, January 10, 2019, compared 3,070 grade 7 through 12 students in Fort McMurray against 2,796 controls in Red Deer, Alberta: nearly six thousand students, with a genuine control population.

From the paper: "37% met conditions for probable diagnosis of PTSD based on thresholding CPSS scores with a cutoff of 15," and "Mental health symptoms were statistically significantly elevated in the Fort McMurray population when compared to the control population in Red Deer." PTSD there is post traumatic stress disorder.

Note the anxiety row: 15 percent versus 16 percent, effectively the same. Report the whole table. A witness who reports only the elevated rows is cherry picking, and under amended Rule 702 that is an exclusion ground, not a cross examination point.

Losing a community rather than a house

Silveira and colleagues, "Chronic Mental Health Sequelae of Climate Change Extremes: A Case Study of the Deadliest Californian Wildfire," International Journal of Environmental Research and Public Health, February 4, 2021, assessed 725 California residents six months after the 2018 Camp Fire. Mean symptom severity: post traumatic stress 26.68 with a standard deviation of 19.50; depression 6.74 with a standard deviation of 5.94; anxiety 7.09 with a standard deviation of 5.65. The conclusion: "direct exposure to large scale fires significantly increased the risk for mental health disorders, particularly for PTSD and depression."

The design is what makes it useful. It compared directly exposed, indirectly exposed, and non exposed groups, and both exposed groups showed elevated depression and anxiety scores. People who did not lose their homes but lost their town, and Paradise lost roughly ninety percent of its structures, were measurably harmed.

The Marshall Fire physical symptom data in chapter fifteen shows the same thing: more than 40 percent of 831 respondents living away from smoke damaged homes reported symptoms, and over half of standing home survivors reported health problems from poor indoor air quality.

But they've been in their own special kind of hell ever since.

Annie Barbour, United Policyholders, 2017 Tubbs Fire survivor, on standing home owners, PBS NewsHour and the Associated Press, January 5, 2026

The time cost, and an honest statement of what is not known

No study or survey quantifies hours spent on wildfire insurance claims, permits, and rebuilding. That number does not exist in published form. Do not invent one, and do not let an expert invent one. What does exist is a set of proxies a jury can understand.

• Twenty two percent of Camp Fire survivors were still working on their contents inventory twenty four months after the fire. Another twenty three percent had given up entirely.
• Forty eight percent of 2025 Los Angeles respondents had three or more adjusters assigned, and forty three percent had to restart the process when adjusters changed, up fourteen points from the six month survey.
• Forty one percent of Los Angeles total loss claimants had not settled the dwelling portion at twelve to thirteen months. Thirty one percent of Marshall claimants had not settled at twenty four months.
• Forty four percent of Los Angeles respondents had not hired professional help of any kind at the one year mark, meaning the labor fell on the household.
• At twelve months, only forty one percent of intended rebuilders had applied for a permit.

The WORST part of all of this is I have been trying to get a permit to rebuild for 170 days and counting. I'm close to giving up!

J.D., Palisades Fire survivor, United Policyholders 2025 Los Angeles Wildfires 12-Month Survey, March 2026

Using this material without exploiting it

Do not diagnose from the podium. Prevalence data establishes that a population was harmed, not that your client has post traumatic stress disorder. A claimed diagnosis needs a treating clinician or a retained forensic psychiatrist, and the treatment records.

Use the eighteen month finding as a causal bridge, not a diagnosis. The published curve shows a second rise at eighteen months and the insurance data shows benefits ending in the same window. Put them side by side and let the jury draw the inference.

In California, know where the emotional distress lives. Under the annoyance and discomfort doctrine from chapter fourteen, emotional distress and mental anguish fold into the property tort recovery, and the restrictions governing negligent infliction of emotional distress do not apply. That is a cleaner path than a standalone distress claim, and every occupant with immediate and personal possession can use it.

Chapter 19

Trying the Case

A wildfire trial fails three ways. The jury does not understand the physics, so causation becomes a credibility contest between two men in suits. The damages case is a spreadsheet, so the jury awards what the adjuster already offered. Or the lawyer overreaches on one element and loses the jury's trust on all of them.

Sequence the physics so the jury never falls behind

Teach in the order the fire happened, not in the order the evidence was gathered. Each step should stand alone, so that if a juror stops following at step four, steps one through three still landed.

1. The day. Weather, wind, humidity, fuel moisture. A jury understands a dry, windy day before it understands anything else, and it is where the notice case lives.
2. The equipment. What it is, what it does, what it is supposed to be inspected for, and what the inspection record says. Show the actual hardware if you can get it admitted.
3. The failure. One mechanism, named and demonstrated. Do not present four alternative mechanisms and let the jury pick. Pick one and explain why the others were eliminated.
4. The ignition. The particle, the fuel, the contact. This is where the numbers from chapter two do their work.
5. The spread. From the ignition area outward, matched to the timeline of 911 calls, satellite detections, and first arrival.
6. The loss. Only now.

What a jury can and cannot absorb

They can absorb one number per exhibit. A slide with three numbers communicates nothing. A slide with one number, said out loud, repeated twice, and never contradicted, becomes a fact the jury owns. Ninety seven percent for the energy release component percentile. One hundred and three degrees for the indicator error. Nine million dollars for the shareholder charge. Two hundred forty seven dollars per square foot for the insurance shortfall.

They can absorb a physical demonstration. A length of conductor with arc erosion on it, passed to the jury box, does more than an hour of testimony. A jar of solidified metal spheres from the origin area is the case in a container. A photograph of a grass stem bent back toward the origin teaches direction of spread in four seconds.

They can absorb a timeline with two rows. What was known and what was done. Not five rows.

They cannot absorb a model output they cannot check. A defense expert will call a fire spread simulation a cartoon, correctly, because the underlying tool is a planning tool and its documentation says so. Use it to illustrate a proposition you have already proved by other means, or not at all.

They cannot absorb a damages spreadsheet. Turn every damages element into a picture. Three stacked bars for the underinsurance gap. One curve for home value by damage status. One map showing the client's parcel inside the debris flow hazard zone.

Building the demonstrative

Everything on it must be independently admissible. A demonstrative that summarizes admitted evidence survives. One that contains a single number the witness cannot source gets excluded whole, and you lose the exhibit for a detail you did not need.

Label the uncertainty on the face of the exhibit. Put "positional uncertainty 375 meters" on the satellite detection map and "planning model, not a forensic reconstruction" on the spread simulation. Doing it yourself makes your weakest point evidence of your candor. Letting opposing counsel do it makes it evidence that you hid something.

Build it to be cross examined. Hand the defense expert your demonstrative and ask him to mark where he disagrees. If the exhibit is honest, that narrows the dispute in front of the jury. If it is not, you find out in preparation rather than at trial.

What moves opposing counsel to settle

Not the size of the demand. Defense counsel and their carriers price cases on risk adjusted exposure, and the demand is an input they discount. Five things move the number.

An admission in a public filing. Read the defendant's securities filings and quote them: "A viable explanation is that a de-energized idle SCE transmission facility in the preliminary area of origin was associated with the ignition of the fire." The defendant wrote that, under penalty, to its own investors. Nothing you write will be more persuasive.

A preservation letter that was ignored. An intent finding under Rule 37(e) is the most expensive thing that can happen to a defendant short of a verdict, and it is within your control to set up on day two.

A damages theory they have not seen before. A defense team that has priced a hundred destroyed home claims has a model. It has none for a surviving home with a fifteen percent viewshed diminution, an arborist's trunk formula appraisal of a lost canopy, and an annoyance and discomfort claim for every occupant on the block. Unpriced theories are expensive to defend.

A bellwether calendar. The Eaton bellwether was set for January 25, 2027 and Edison's request to move it to August was rejected. That ruling was worth more than a hundred pages of briefing.

A fee provision. In California, inverse condemnation carries attorneys' fees and costs under Code of Civil Procedure section 1036 and a negligence claim does not. That asymmetry is a settlement driver and both sides know it.

The three concessions to make early

Concede that the weather was extraordinary. Fighting it makes you look unreasonable and does not help, because the notice case depends on the conditions being extreme and foreseeable at the same time.

Concede the limits of the origin methodology. Say the error rate out loud in your case in chief and explain that the origin opinion does not rest on a single artifact. A jury forgives a limitation that is disclosed and punishes one that is exposed.

Concede that some damages are hard to measure, then show your work on the ones that are not: this element I can prove to the dollar, this one within a range, and this one I am not asking you for. That buys credibility for the whole schedule.

My own view, from the witness stand

Juries do not have trouble with physics. They have trouble with experts who will not admit what they do not know. I have watched extremely qualified people lose a room by refusing to say "I do not know" once. Concede a genuine limit and the jury starts believing everything else the witness says. Defend an indefensible detail and they start auditing every number he gave them.

The damages case is where these trials are won, and almost nobody prepares it like it matters. Liability is frequently close to conceded by the time you reach trial, or decided by a doctrine the jury never hears about. What the jury decides is what it was worth. Spend eighty percent of your preparation on causation and twenty on damages and you have prepared the part that is not in dispute.

The last thing is the one this book was written for. Walk the burn scar, not the origin area but the whole thing. Walk down the street where six houses out of ten are gone and four are standing, and knock on the four doors. Those four families have a case nobody has told them about. Their insurance company has closed their file. Their neighborhood is not coming back for a decade. Their property is worth ten to sixteen percent less than it was the day before the fire. They are the losses nobody claims. Claim them.

Appendix A

Glossary

Advanced metering infrastructure last gasp

A final signal transmitted by a smart meter as it loses power, useful for timing a circuit failure to the second.

Arc mapping

In structure fires, locating arc sites on conductors to infer fire travel. In wildfire, locating the arc or fault point on a circuit from physical arc erosion plus relay derived fault distance.

Backing fire

Fire spreading into the wind or downslope, the slowest and least intense mode, leaving the most residual fuel. The origin is normally near the heel.

Combustion byproducts of concern

Industry term for the five contaminant classes in wildfire smoke damage: soot, char and ash; volatile organic compounds; semi volatile organic compounds; heavy metals; and asbestos.

Demand surge

The post catastrophe increase in construction costs, commonly measured at fifteen to thirty percent, with little evidence that prices meaningfully recede afterward.

Difference in differences

An econometric method comparing changes in an affected area against changes in a similar unaffected benchmark region.

Energy release component

The available energy in British thermal units per square foot within the flaming front at the head of a fire. An open ended relative scale that excludes wind and slope.

Fire pattern indicator

A physical object displaying fire effects that can reveal the direction of fire progression at a precise location with accurate analysis.

Foehn wind

A warm, dry downslope wind produced when air descends a lee slope and warms adiabatically while its absolute moisture stays constant. Regional names include Chinook, Santa Ana, Diablo, Sundowner, and East wind.

Haines index

The lower atmospheric severity index, the sum of a stability term and a moisture term each scored one to three, estimating the potential for large plume dominated fire growth.

Joint evidence examination protocol

A written agreement governing who may examine physical evidence, in what sequence, with what documentation, and whether destructive testing is permitted.

Macro and micro indicators

Macro indicators are large object fire effects such as shrubs, trees, and crown scorch. Micro indicators are small object effects such as grass stems and pebbles. These terms modify indicators, not origin areas.

NFPA 921

The National Fire Protection Association Guide for Fire and Explosion Investigations. A guide, not a standard. Chapter 27 covers wildfire.

NFPA 1033

The National Fire Protection Association Standard for Professional Qualifications for Fire Investigator, including the sixteen required knowledge areas.

PMS 412

The National Wildfire Coordinating Group Guide to Wildland Fire Origin and Cause Determination, March 2025 version, the federal wildland analogue to NFPA 921.

Public safety power shutoff

Preemptive de-energization of circuits during extreme fire weather. The decision record is discoverable and important.

Recloser

A protective device that automatically re-energizes a circuit after a fault, on the assumption the fault was transient. Re-energizing a downed conductor produces a second arcing event.

Spotting

Firebrand lofting and downwind ignition of a receptive fuel bed. A spot fire creates its own local origin with its own indicator field.

Supervisory control and data acquisition

The utility's operational monitoring and control system. Its time series is the highest value electronic evidence in a utility ignition case.

Timelag class

Fuel classification by response time to environmental moisture: 1 hour, 10 hour, 100 hour, and 1000 hour. Fine fuels respond within an hour.

Trunk formula technique

The cost approach method in the Council of Tree and Landscape Appraisers Guide for Plant Appraisal for appraising large specimen trees, extrapolating nursery stock cost to the size of the appraised tree and depreciating for condition and limitations.

Wildfire Fund

The approximately twenty one billion dollar California fund created by Assembly Bill 1054 to reimburse investor owned utilities for third party wildfire damages above a threshold.

Appendix B

Sources by chapter

Chapters 2, 3, and 4: ignition, origin, and method

• National Fire Protection Association, NFPA 921 product page: nfpa.org
• National Fire Protection Association, NFPA 1033 product page: nfpa.org
• National Wildfire Coordinating Group, PMS 412: nwcg.gov/publications/pms412
• National Wildfire Coordinating Group course FI-110 Unit 4, fire pattern indicators: training.nwcg.gov
• National Wildfire Coordinating Group course FI-110 Unit 5, cause categories: training.nwcg.gov
• Bureau of Indian Affairs, Wildland Fire Origin and Cause Investigation Handbook, 90 IAM 6-H: bia.gov
• Bureau of Land Management, how to investigate a wildland fire: blm.gov
• Urban, Zak and Fernandez-Pello, spot fire ignition by hot aluminum particles, Fire Technology 54(3), 2018: frg.berkeley.edu
• Simeoni and colleagues, fire pattern indicator field experiment, Journal of Fire Sciences 35(5), 2017: fs.usda.gov
• Parker and Babrauskas, validation of directional indicators, Fire 2024, 7(1), 5: mdpi.com
• International Association of Wildland Fire, fire pattern indicators position article: iawfonline.org
• Texas A&M Engineering Experiment Station, how power lines cause wildfires: wildfiremitigation.tees.tamus.edu
• Vaisala National Lightning Detection Network: vaisala.com

Chapter 5: weather, fuel, and public data

• National Wildfire Coordinating Group PMS 437, dead fuel moisture: nwcg.gov
• Energy release component: wildfire.gov
• National Wildfire Coordinating Group, atmospheric stability and the Haines index: nwcg.gov
• National Weather Service Denver and Boulder, red flag warning definitions: weather.gov/bou
• National Wildfire Coordinating Group, critical fire weather: nwcg.gov
• Western Regional Climate Center remote automated weather station archive: raws.dri.edu
• FAMWEB Data Warehouse: wildfire.gov
• Iowa Environmental Mesonet: mesonet.agron.iastate.edu
• NASA Fire Information for Resource Management System: firms.modaps.eosdis.nasa.gov
• National Interagency Fire Center statistics: nifc.gov
• FARSITE technical limitations and assumptions: owfflammaphelp62.firenet.gov

Chapter 6: evidence and spoliation

• California Public Utilities Commission General Order 95, Rule 19: ia.cpuc.ca.gov
• California Public Utilities Commission wildfire staff investigations: cpuc.ca.gov
• Pacific Gas and Electric letter to the Safety and Enforcement Division, October 13, 2017: cpuc.ca.gov
• Federal Rule of Civil Procedure 37(e): law.cornell.edu
• Bonneville Power Administration spoliation ruling reporting: opb.org
• Department of Forestry and Fire Protection v. Howell opinion: courts.ca.gov

Chapter 7: admissibility

• Federal Rule of Evidence 702, current text: law.cornell.edu
• United States v. Idaho County Light and Power Cooperative Association: govinfo.gov
• Cozen O'Connor, fire experts and NFPA 921 under Daubert: cozen.com
• Babrauskas, arc mapping critique: doctorfire.com
• May and Icove, arc mapping methodologies, 7 Lincoln Memorial University Law Review 38: digitalcommons.lmunet.edu
• Washington Legal Foundation, amended Rule 702 in the circuits: wlf.org

Chapters 8 and 9: liability and cost allocation

• Barham v. Southern California Edison: findlaw.com
• Simple Avo Paradise Ranch v. Southern California Edison: findlaw.com
• California Public Utilities Commission Decision 17-11-033: docs.cpuc.ca.gov
• Assembly Bill 1054 text: leginfo.legislature.ca.gov
• California Public Advocates Office, ratepayer impacts of strict liability and inverse condemnation: publicadvocates.cpuc.ca.gov
• James v. PacifiCorp, Oregon Court of Appeals, April 8, 2026: justia.com
• PacifiCorp wildfire litigation page: pacificorp.com
• Stanford Law, changing utility wildfire liability laws: law.stanford.edu
• Pacific Gas and Electric Fire Victim Trust: firevictimtrust.com
• Hawaii Supreme Court, In re Maui Fire Cases: justia.com

Chapter 10: criminal exposure

• Butte County District Attorney, Camp Fire Public Report, June 16, 2020: buttecounty.net
• North State District Attorneys, Dixie Fire settlement, April 11, 2022: buttecounty.net
• Department of Justice, San Bruno sentencing: justice.gov
• California Penal Code section 452: justia.com
• California Penal Code section 1202.4: justia.com
• California Health and Safety Code section 13009: justia.com
• California Civil Code section 3333.8, prescribed burning: justia.com
• Forest Service, Gallinas and Las Dispensas prescribed fire review: fs.usda.gov
• Department of Justice, PacifiCorp $575 million settlement: justice.gov
• Department of Justice, Southern California Edison Bobcat Fire settlement: justice.gov
• United States v. Jonathan Rinderknecht: justice.gov

Chapter 11: the Marshall Fire

• Boulder County Sheriff's Office cause and origin conclusion: bouldercounty.gov
• Boulder County damage assessment: bouldercounty.gov
• Boulder County Marshall Fire recovery: bouldercounty.gov
• Public Service Company of Colorado Form 10-K for fiscal 2025: sec.gov
• United Policyholders on Marshall Fire underinsurance and estimating software: uphelp.org
• Boulder Reporting Lab, standing home survivors: boulderreportinglab.org

Chapter 12: Eaton and Palisades

• CAL FIRE Palisades Fire incident page: fire.ca.gov
• CAL FIRE Eaton Fire incident page: fire.ca.gov
• California Department of Insurance claims tracker release: insurance.ca.gov
• Edison newsroom, direct compensation program metrics: newsroom.edison.com
• Edison rebuttal to media coverage, June 10, 2026: newsroom.edison.com
• California Attorney General civil rights investigation: oag.ca.gov

Chapters 13 and 14: property damages, trees, view, and stigma

• California civil jury instruction 3903F, damage to real property: justia.com
• California civil jury instruction 2002, trespass to timber: justia.com
• California civil jury instruction 2031, annoyance and discomfort: justia.com
• Heninger v. Dunn: justia.com
• Kelly v. CB&I Constructors: findlaw.com
• Scholes v. Lambirth Trucking Co.: findlaw.com
• Fulle v. Kanani: findlaw.com
• Stetler, Venn and Calkin, wildfire proximity and property value: fs.usda.gov
• Mueller, Loomis and González-Cabán, repeated wildfire and house prices: springer.com
• Federal Reserve Bank of San Francisco, wildfires and real estate values: frbsf.org
• Benson and colleagues, the value of a view: sauder.ubc.ca
• Council of Tree and Landscape Appraisers trunk formula technique explained: tcimag.tcia.org
• Komen, field precision of the trunk formula method: jameskomen.com
• Drake Law Review, stigma damages: drakelawreview.org
• American Bar Association, mechanisms underlying property diminution damages: americanbar.org
• Horvitz & Levy, damages in wildfire litigation: horvitzlevy.com
• Santa Monica Mirror report of the Realtor.com and Cotality analysis: smmirror.com
• Oregon Revised Statutes 105.810: oregon.public.law
• Washington timber trespass statute discussion: beresfordlaw.com

Chapters 15 and 16: smoke, debris flow, and agriculture

• IICRC, RIA and CIRI Technical Guide for Wildfire Restoration, Version 2: insurance.ca.gov
• ANSI/IICRC S700 publication announcement, with the wildfire exclusion: cleanfax.com
• Standing homes survivor letter to the California task force: insurance.ca.gov
• National Collaborating Centre for Environmental Health, indoor smoke residues review: ncceh.ca
• United Policyholders smoke claims webinar deck: uphelp.org
• United Policyholders on the Aliff ruling: uphelp.org
• California Department of Insurance Bulletin 2025-7 release: insurance.ca.gov
• California Department of Insurance action against the FAIR Plan: insurance.ca.gov
• California Department of Insurance action against State Farm: insurance.ca.gov
• Smoke Damage Recovery Act announcement: insurance.ca.gov
• United States Geological Survey post-fire debris flow hazards: usgs.gov
• Kean and colleagues, the January 9, 2018 Montecito debris flow, Geosphere: geoscienceworld.org
• Kropp and De Andrade, wildfire smoke exposure and the wine industry, Choices: choicesmagazine.org
• Aaron Smith, assessing fire damage to the 2020 wine grape crop: asmith.ucdavis.edu

Chapters 17 and 18: insurance and the human cost

• United Policyholders survey index: uphelp.org/media/surveys
• United Policyholders 2025 Los Angeles wildfires 12 month survey: uphelp.org
• United Policyholders 2021 Marshall Fire 12 month survey: uphelp.org
• United Policyholders 2021 Marshall Fire 24 month survey: uphelp.org
• Klein, "Truth and Consequences," Lewis & Clark Law Review Vol. 30.1, 2026: uphelp.org
• California FAIR Plan key statistics: cfpnet.com
• Treasury Federal Insurance Office, homeowners insurance markets 2018 to 2022: uphelp.org
• Biswas, Hossain and Zink, California wildfires, property damage, and mortgage repayment: uphelp.org
• To, Eboreime and Agyapong, the impact of wildfires on mental health: mdpi.com
• Brown and colleagues, Fort McMurray student mental health: bmcpsychiatry.biomedcentral.com
• Silveira and colleagues, chronic mental health sequelae of the Camp Fire: mdpi.com
• Consumer Federation of America on uninsured homeowners: consumerfed.org

Appendix C

Verification punch list before publication

Every item below is used in this book with a hedge, or was excluded from an assertion, because it could not be confirmed to a primary source during research. Nothing here is an apology. It is the list of documents to pull before any of this goes into a brief, a report, or a second edition.

Priority one: the numbers that carry chapters

1. The untouched homes valuation table, chapter fourteen. Reached second hand through the Santa Monica Mirror's report of a Realtor.com analysis of Cotality and Quantarium data. The primary Realtor.com piece was not retrievable. The percentage column is arithmetic performed on reported dollar figures, not published percentages. This is the headline dataset of the centerpiece chapter. Get the primary analysis.
2. The 74 percent and $139,000 Marshall underinsurance pairing. Pin it to its underlying study. United Policyholders elsewhere states roughly two thirds, and its own twelve month survey reports 80 percent with a shortfall of $110 per square foot.
3. The multi year debris flow risk window, chapter sixteen. The United States Geological Survey overview page supplies no number. The likely primary source is the Geosphere paper on characterizing changes in post-fire debris flow hazard as burned areas recover. Get the numeric window.
4. Photographs. No photographs appear in this book. Wikimedia Commons was unreachable during production, so no image URL, license, or file existence could be verified. Every figure in this book is an original drawing. If photographs are added in a second edition, verify each file description page, the license, and the direct file URL before use.

Standards and technical documents

5. NFPA 921 (2024) effective date. The preview front matter renders it as January 10, 2023, which is almost certainly a typographical error for 2024.
6. NFPA 921 (2024) pinpoint subsection numbers for expectation bias, confirmation bias, and levels of certainty. Chapter 4 and Chapter 19 top level headings are confirmed; the sub-subsections are not.
7. NFPA 1033: whether the 2026 edition has been issued, and the section number of the sixteen knowledge areas in the 2022 edition. The 2014 number is 1.3.7 and the pending revision moves it to 4.1.7.
8. PMS 412 (March 2025) chapter and section structure. The file exceeded fetch limits. All PMS 412 derived content in this book comes from National Wildfire Coordinating Group course FI-110 instructor guides, the PMS 205 glossary, and the Bureau of Indian Affairs handbook, which restate it.
9. A verbatim PMS 412 definition of "die-out pattern." It is standard field vocabulary but no verbatim definition was retrieved.
10. The Parker and Babrauskas characterization conflict. The International Association of Wildland Fire article describes the 2024 paper as validating investigator reliability; the paper's own conclusion is the opposite. Read both primary sources word for word.
11. Catalytic converter surface temperature versus vegetation ignition. No quantitative data was located.
12. McAllister and Finney fire safe cigarette ignition probabilities. Full text not retrievable; pull directly for the numbers.
13. Lithium-ion and battery energy storage system ignition. Entirely unresearched. Needed before any battery ignition opinion: thermal runaway as a wildland ignition source, the physical signature, the distinction between a battery that ignited a fire and one consumed by it, and the applicable National Fire Protection Association section.

Cases, dockets, and citations

14. Holiday Farm and Bonneville Power Administration spoliation ruling: docket number and official citation.
15. Department of Forestry and Fire Protection v. Howell: official California Appellate Reports citation, believed to be 18 Cal. App. 5th 154.
16. Chapman v. State Farm Fire & Casualty Co.: full citation.
17. California expert gatekeeping: Sargon Enterprises, Inc. v. University of Southern California, 55 Cal. 4th 747 (2012), and Evidence Code sections 801(b) and 802. Cited in the research from background knowledge and not verified.
18. Santa Fe Partnership v. ARCO Products Co. (1996) 46 Cal.App.4th 967: the specific holding was not extracted. It is the leading California obstacle to a residual stigma theory.
19. Marshall Fire coordinated proceeding: court, case caption, case number, judge, whether a class was certified, and the theories pleaded.
20. Palisades Fire litigation against the City of Los Angeles and its Department of Water and Power: consolidation structure, judge, rulings, trial date, and the status of inverse condemnation claims against a municipal utility.
21. Eaton Fire coordinated proceeding: the Judicial Council coordination number, and separately itemized subrogation and public entity settlement amounts, which Edison aggregates deliberately.
22. PacifiCorp's published verdict table contains non sequential trial numbers and an internally inconsistent reference to an argument date that predates the April 8, 2026 opinion. Verify against the docket.
23. Texas Attorney General filing date discrepancy: the press release says December 2024, contemporaneous reporting says December 16, 2025. Check the Hemphill County docket.
24. Camp Fire indictment statute sections. The public report does not print them. Conventionally Penal Code sections 192(b) and 452, but the subdivisions are unconfirmed.
25. Outcome of the January 15, 2021 Camp Fire restitution hearing.
26. The Zogg Fire court quotation on standard of care: the judge is unnamed in the available source.
27. Caldor Fire: reporting indicates all charges against one defendant were dropped in April 2024. Unconfirmed.

Facts and figures

28. Marshall Fire peak gust of 115 miles per hour: the instrument and station location.
29. Marshall Fire snow drought day count before December 30, 2021.
30. Marshall Fire fatalities: names and ages.
31. The CAL FIRE Palisades Fire start time conflict: 2:30 a.m. on the incident page versus 10:30 a.m. as widely reported. Do not print either until resolved.
32. East Altadena evacuation order clock time, commonly reported as about 7:26 p.m. January 7, and the minute level West Altadena time, commonly 3:25 a.m.
33. Whether a Los Angeles County after action review of the Eaton Fire response exists. If it does, it is the central document for that chapter.
34. The Los Angeles Department of Water and Power former chief engineer quoted on the system design basis: name not captured.
35. Los Angeles Fire Department Chief Kristin Crowley's memorandum dated December 4, 2024 on budget and overtime cuts. Verify the primary document. Its timing, five weeks before the fires, makes it potentially the most valuable single document in that chapter.
36. Hermit's Peak and Calf Canyon administrative overhead, reported separately at about $377 million.
37. Union Pacific, Storrie Fire, reported at $102 million in 2008. Source paywalled.
38. Aggregate civil settlements to individual victims for the Dixie and Zogg fires. Not publicly aggregated; settled confidentially.
39. Davey Tree settlement amount with the Fire Victim Trust, and the final trust pro rata percentage.
40. Whether any Maui settlement funds have actually reached victims.
41. Whether any Colorado statute limits utility wildfire liability. None was found. Confirm the negative before asserting it.
42. Colorado FAIR Plan coverage limits, eligibility detail, and policy counts. Not published on its site.
43. State by state non renewal rates. Only risk tier breakdowns were obtained.
44. Completed rebuild counts for Paradise, the Marshall Fire communities, and Altadena. Los Angeles County data sits in a dashboard that could not be extracted; the data contact is the county rebuild office.
45. Any quantified study of hours spent on wildfire claims, permits, and rebuilding. None is believed to exist in published form. Do not invent one.
46. Completed suicide data among wildfire survivors. Not found in the peer reviewed scoping review. Only suicidal ideation is solidly sourced.
47. Two USDA Forest Service tree canopy meta-analyses on the implicit value of tree cover. Both were inaccessible; the headline percentages are missing from this book entirely and should be added.
48. The California Department of Insurance Smoke Claims and Remediation Task Force final report of March 2026. Two appendices were retrieved; the parent report was not.
49. State prescribed burn liability standards. The 2006 four way survey cited here is stale. Re-survey state by state.
50. The Doug Boxer statement that direct compensation applicants will likely receive about half of what they are owed is a paraphrase in the source. Confirm the original wording before quoting it.
51. Two Commissioner Lara quotations from the December 13, 2024 catastrophe modeling release came back flagged as possibly paraphrased. Neither is used in this book. Re-read the release before using them.
52. Southern California Edison compensation program figures conflict across the July 16, 2026 press release, the July 30, 2026 earnings call, and trade press. Cite the press release and label offered versus paid precisely.

Appendix D

About the author

Steve Wolf is an expert witness in firearms, pyrotechnics, stunts, fire, on-set safety, wildfire, theatrical rigging, self contained underwater breathing apparatus, and explosion cases. He has spent his career in production, building, firing, rigging, and detonating the systems he now testifies about. He founded RangeMaster, Tactical Choices, Wolf Stuntworks, Pyro School, Team Wildfire, Wildfire Labs, Stunt Ranch, and Science in the Movies. He is a patented inventor of fire suppression technology. He holds a Federal Firearms License, a Federal Explosives Manufacturing License, and certifications in emergency management, incident command, active shooter response, wildfire management, and mass casualty response. He is a Columbia University graduate. He won a $66.5 million personal injury verdict in New Mexico. He is retained by attorneys nationwide, on both sides, in civil and criminal matters.

On wildfire specifically: Team Wildfire is a fire management and suppression technology company he founded. Wildfire Labs, which he also founded, is a nonprofit developing and demonstrating aerial fire suppression technology. His fire suppression inventions include Hurricane, a jet engine firefighting system that atomizes and projects retardant on high velocity exhaust; WolfPack, a backpack micro jet turbine suppression unit; and Wildfire Command, a real time resource allocation engine, United States Patent Application 18/630,213. Others are CoreTechs Helmet, ThermaFlow-TankGuard, PyroSphere Shelter, BioSafe FireGuard, Cyclone Express, Humidi-Fire, SkyPhoon, and Sand Storm. He holds emergency management certifications in wildfire mitigation and incident command, and he has tried a wildfire property loss case, Kingstone v. Doe, for the plaintiff, on real estate value lost to wildfire, in Boulder County, Colorado.

Available to attorneys nationwide. SteveWolfExpertWitness.com