The Case for Preparedness

Wildfire · Lahaina, Hawaii · August 2023

Documented August 2026

The Lahaina Fire

A wind-driven wildfire, worsened by drought conditions and gusts from a passing hurricane, tore through the historic town of Lahaina on August 8, 2023.

What made Lahaina a case study in system failure, not just fire: Maui County's extensive outdoor warning siren network, one of the largest in the world, was never activated. Hawaii Attorney General Anne Lopez's official review, along with a separate investigation by the Western Fire Chiefs Association, documented a cascade of breakdowns: officials relied on cell phone alerts and social media posts, but widespread power and cellular outages meant many residents never received them. Survivors interviewed at evacuation centers said their first warning was the sight of flames or the sound of explosions nearby, not any official alert. Evacuation routes were limited and became gridlocked; some people died in their cars.

The initial damage assessment put the cost of rebuilding at $5.5 billion; a year later, once debris removal, environmental remediation, temporary housing, and the actual work of rebuilding were counted, the full recovery cost was estimated to exceed $12 billion, more than double the first number, and a pattern that repeats across nearly every disaster on this list.

Do this: confirm your phone actually receives Wireless Emergency Alerts and get a battery- or crank-powered NOAA weather radio that works when cell towers and power both fail, exactly what Lahaina's residents didn't have. See Family Communication Plan & Emergency Radio.

Grid failure · Texas · February 2021

Documented August 2026

Winter Storm Uri

A severe winter storm pushed Texas's power grid to failure in February 2021. ERCOT, the state's grid operator, ordered roughly 20,000 megawatts of rolling blackouts, the largest manually controlled load-shedding event in U.S. history, leaving more than 4.5 million people without power, some for four days in subfreezing temperatures. Frozen wind turbines, natural gas infrastructure that lost pressure in the cold, and even a nuclear plant sensor failure combined to take generation offline exactly when demand peaked.

The Texas Department of State Health Services eventually attributed 210 deaths to the storm, most from cold exposure, though earlier tallies had been far lower, a reminder that official death counts from a disaster like this often rise substantially after the initial reporting. Water systems failed alongside the power grid: frozen and burst pipes left millions under boil-water notices for days.

Do this: know your household's actual backup heat source before a real freeze, not during one, and leave a faucet dripping in the coldest part of your house to keep pipes from bursting. See Staying Warm Without Power.

Inland flooding · Western North Carolina · September 2024

Documented August 2026

Hurricane Helene

Helene is a useful case specifically because western North Carolina is mountainous and inland, not the coastal flood zone most people picture when they think "hurricane." The storm's remnants dropped catastrophic rainfall on terrain that funneled it into flash floods, killing at least 68 people in North Carolina alone (with the toll still rising as of official reporting) and causing an estimated $59.6 billion in damage and recovery needs statewide, according to North Carolina's Office of State Budget and Management.

The communications failure was the most severe on record for a U.S. disaster: high winds and floodwaters destroyed more than 1,700 miles of fiber-optic cable, damaged roughly $100 million in communications infrastructure, and cut off 200,000 internet and phone customers, isolating an estimated 150,000 people from every layer of the emergency communications system (cell, landline, internet, and broadcast) at once.

Do this: pick an out-of-area contact today and write their number down on paper, since Helene proved cell, landline, internet, and broadcast can all fail together, at once, with no layer left as a fallback. See Family Communication Plan & Emergency Radio.

Wildfire · Boulder County, Colorado · December 2021

Documented September 2026

The Marshall Fire

On December 30, 2021, hurricane-force winds, gusting past 100 mph, drove two separate grass fires together into a single blaze that tore through the suburbs between Denver and Boulder. In about six hours it destroyed 1,084 homes, damaged 149 more, and burned 30 commercial structures, killing two people and forcing roughly 37,500 people to evacuate with little warning.

What makes Marshall a useful case study is how badly it breaks the mental picture most people have of "wildfire." This wasn't a rural forest burning in August; it was dense, ordinary subdivisions burning in winter, after an unusually wet spring grew heavy grass that an unusually dry autumn then cured into fuel. Neighborhoods with paved streets, sidewalks, and two-car garages burned block by block, the same way Lahaina's downtown did: dense residential construction with wood-frame homes and closely spaced lots carries fire from house to house just as readily as timber carries it tree to tree.

For a Wasatch Front reader, Marshall is arguably the closest analog on this list: suburban wildland-urban interface fire, the exact terrain of the benches above Salt Lake City, Bountiful, and Cottonwood Heights, not the remote forest and rangeland of Lahaina or the Cottonwood Fire later in this section.

Do this: stage a go-bag by your main exit today, Marshall gave residents six hours from ignition to over 1,000 homes destroyed, not enough time to pack from scratch. See Go-Bags & 72-Hour Kits.

Wildfire, government-caused · San Miguel & Mora counties, New Mexico · April–August 2022

Documented September 2026

The Hermits Peak/Calf Canyon Fire

On April 6, 2022, a U.S. Forest Service prescribed burn in the Santa Fe National Forest, meant to reduce wildfire risk, escaped control in unexpected winds. A second prescribed-burn holdover ignited nearby three days later and merged with the first. Together they burned for four months and consumed 341,471 acres, the largest wildfire in New Mexico's recorded history, killing 4 people, destroying at least 903 structures, and forcing more than 27,000 people to evacuate.

Hermits Peak belongs on this list for a reason none of the other fires here share: the government itself started it, trying to prevent exactly this outcome. Congress responded with the Hermit's Peak/Calf Canyon Fire Assistance Act, appropriating $2.5 billion in September 2022 and a further $1.5 billion in December 2024, for a total of $5.45 billion. As of March 2025, the FEMA claims office had paid out $1.97 billion, a little over a third of that total, nearly three years after the fire started, a reminder that even well-funded federal compensation moves far slower than the disaster it's paying for.

The fire's aftermath introduced a hazard most wildfire coverage skips: with the watershed's vegetation burned away, monsoon rains have repeatedly sent flash flooding and debris down the burn scar, killing at least three more people in the weeks after the fire itself was contained, and washing ash and sediment into the Gallinas River, the primary water source for Las Vegas, New Mexico, a city of 12,000. Water officials say the risk to the city's drinking water will last five to ten years, until the burned hillsides regrow enough to hold soil and moisture again. In January 2025, nearly three years after the fire, contamination got bad enough to shut off the city's water supply entirely. It's the same fire-then-water-crisis sequence as the Cottonwood Fire that closes out this section, playing out at a much larger scale and for years longer, because the burn scar behind it was orders of magnitude bigger.

Do this: photograph and document what you own now, before you need to prove it. Hermits Peak's own compensation fund took nearly three years to pay out a third of what was owed. See Financial Preparedness Basics.

Ransomware · U.S. East Coast · May 2021

Documented August 2026

The Colonial Pipeline Attack

Not every emergency is a storm. On May 7, 2021, the ransomware group DarkSide encrypted the IT systems of Colonial Pipeline, which carries more than 100 million gallons of gasoline, diesel, and jet fuel a day from Texas to New York. The company shut the pipeline down as a precaution and didn't fully resume service for five days.

The physical damage was zero: no fuel was lost, no infrastructure was destroyed, and no one was hurt. The disruption was entirely about the response: panic buying emptied gas stations across the Southeast (more than 70% of stations in some metro areas ran dry), some drivers filled plastic bags with gasoline, and the average pump price rose six cents a gallon in a week, the sharpest jump since 2014. Georgia, Virginia, North Carolina, and Florida all declared states of emergency over a shortage that started with zero fuel actually being destroyed. The lesson isn't about pipelines specifically; it's that a single point of failure in infrastructure most people never think about can turn into a real, if temporary, shortage anywhere it touches, with almost no warning.

Do this: make a habit of refueling once your tank drops below half, so a run on the pumps, not just a physical shortage, never catches you empty. See Vehicle Emergency Kit & Roadside Basics.

Energy security & conflict · Strait of Hormuz · February 2026–ongoing

Documented September 2026

The Strait of Hormuz Closure

On February 28, 2026, the United States and Israel launched a military campaign against Iran that included the assassination of Iran's supreme leader. Within 48 hours, Iran's response, naval mines, small-boat attacks, and the seizure of merchant vessels, had effectively shut the Strait of Hormuz, the waterway carrying roughly one-fifth of the world's oil supply. Daily transits collapsed from around 100 ships in February to as few as six a day in early March, as shipping insurers made coverage for the route unavailable or prohibitively expensive.

The toll has been direct, not just economic: at least 19 mariners have been killed, along with a port worker in Bahrain; at least one tugboat sunk, 17 or more merchant ships damaged and seven abandoned, and two vessels captured. At the crisis's peak in May 2026, roughly 1,550 vessels and 22,500 mariners were stranded in or near the Gulf. A ceasefire briefly reopened the strait in early April, but attacks resumed and have continued as recently as August 2026, more than five months after the crisis began.

Global oil markets absorbed one of the largest single oil-market shocks on record: Brent crude jumped roughly 65% by the end of March and peaked near $126 a barrel in early May. The impact reached American households directly: by June 1, U.S. regular gasoline had climbed to $4.31 a gallon and diesel to $5.35, even though the U.S. is less dependent on Hormuz shipping than Asian and European buyers and was expected to feel the impact last and least.

Do this: keep a real cash reserve and know your generator's actual fuel consumption, a shock like this arrives with no storm to warn you and no local event to point to. See Financial Preparedness Basics and Sizing a Fuel-Powered Generator.

Earthquake · Salt Lake Valley, Utah · March 2020

Documented August 2026

The Magna Earthquake

This is one of two entries on this list that happened here; the other, a wildfire followed by a flood, closes out this section next. On March 18, 2020, a magnitude 5.7 earthquake centered near Magna, Utah struck the Salt Lake Valley: the first earthquake of its size to hit the valley since Salt Lake City was founded, and the strongest in Utah since 1992. It knocked out power to nearly 50,000 homes and businesses and forced a temporary closure of Salt Lake City International Airport. Over 2,500 aftershocks followed.

Magna was, by seismic standards, a moderate event, and it still closed an airport and cost over half a billion dollars. That nobody died is a matter of timing, not preparedness: Utah schools had just suspended in-person learning because of COVID-19, so classrooms were empty when the quake hit at 7:09 a.m. West Lake STEM Junior High in West Valley City was damaged badly enough to be declared a total loss; had students been inside, the outcome would have been very different. It is also, per the state's own seismologists, a preview: the Utah Geological Survey has said explicitly that Magna was within the range of expected background seismicity, not evidence that "the big one" has been released. The real risk profile for this region is the subject of Part Two.

Do this: strap your water heater and anchor heavy furniture this week, Magna was a moderate event and it still cost $629 million; the next one won't be timed to an empty school building. See Securing Your Home Against Earthquake Damage.

Wildfire & flash flooding · Beaver County, Utah · June–July 2026

Documented September 2026

The Cottonwood Fire and the Flood That Followed It

This is the most recent entry on this list, and it's still unfolding as of this writing. On June 22, 2026, the Cottonwood Fire ignited east of Beaver, Utah, from an undetermined human cause. It spread fast, topping 60,000 acres within two days, and by 95% containment in mid-July had burned 97,464 acres across Beaver and Piute counties, Utah's sixth-largest wildfire on record, destroying at least 150 structures.

The fire alone would have been a serious story. What made it a case study in cascading risk is what happened three weeks later: on July 19, more than an inch and a half of rain fell on the Cottonwood burn scar in just 45 minutes. With the vegetation and root systems that normally hold mountain soil in place already burned away, the runoff came down as a debris flow instead of a stream, tearing out sections of State Route 153 through Beaver Canyon so badly UDOT closed it indefinitely, and reaching the town of Beaver, where floodwaters damaged roughly 200 homes and shut down Interstate 15 for several hours. The same flood contaminated Beaver City's municipal water supply, triggering a boil-water advisory that lasted weeks.

Governor Spencer Cox's resulting state of emergency for Beaver, Piute, and Sevier counties covered the wildfire and the flood as a single event, because that's what they were. A burned mountainside doesn't stop being a hazard once the flames are out; for months or years afterward, it's a flood risk instead, and the two disaster types this document treats separately elsewhere in Part One, fire (Lahaina, Marshall) and flood (Helene), played out here as one continuous emergency, in the same county, three weeks apart.

Do this: if a fire has already passed near you, treat the burn scar above your home as an active flood risk for years afterward, not a closed chapter, and know how to treat water if the municipal supply is contaminated again. See Water Purification.

Water contamination · Flint, Michigan · 2014–present

Documented August 2026

The Flint Water Crisis

In April 2014, facing a budget shortfall, a state-appointed emergency manager switched Flint's water source from Detroit's system to the Flint River to save money. The river water wasn't treated with corrosion control, and lead leached from the city's aging pipes into the water supply of a city the state itself had decided needed emergency financial oversight.

The CDC estimates nearly 100,000 Flint residents were exposed to lead-contaminated water; public health estimates put 6,000 to 12,000 children at elevated blood lead levels, a metal with no safe threshold for a developing child. A concurrent Legionnaires' disease outbreak, linked to the same water source switch, killed at least 12 people and sickened roughly 90 more between June 2014 and October 2015.

The state settled the resulting lawsuits for $626 million in 2021, including an agreement to replace every lead service line at no cost to residents. A decade after the switch, roughly 4,000 homes still had lead pipes, according to a 2024 court progress report, a reminder that even a well-funded fix takes years to reach every household once both trust and infrastructure have broken.

Do this: get your own water tested rather than trusting a municipal announcement to reach you in time, Flint's own lead crisis took over a year of resident pressure before officials confirmed what was already true. See Water Testing: Strips, Labs & Field Methods.

Grid failure · Gary, Indiana · August 2026

Documented September 2026

The Gary Derecho Outage

On August 11, 2026, a derecho with winds reaching 99 mph, comparable to a Category 2 hurricane, tore through Northwest Indiana, uprooting trees and downing power lines across Gary and Lake County. At its peak, the storm knocked out power to as many as 374,500 NIPSCO customers, what the utility itself called the largest outage event in its history.

Full restoration in Gary took nearly two weeks, a 14-day outage that ranks among the longest in modern U.S. history for a single storm event, not a multi-state hurricane, while ComEd, serving the Illinois side of the same storm just a few miles away, restored nearly all of its customers within days. Gary is a predominantly Black city, and residents, local leaders, and the Gary NAACP pointed to a visible reason for the gap: the city's grid is old and runs almost entirely above ground through tree-lined streets, while wealthier, predominantly white suburbs nearby have already converted to modern underground lines that don't go down when a tree falls.

What makes Gary a case study in institutional neglect, not just bad luck, is what came out once people started asking why the grid failed this badly. NIPSCO had been authorized to spend nearly $77 million on vegetation management between 2023 and 2025 and more than $2 billion on infrastructure upgrades since 2016, spending residents say never showed up on their own streets. A lawsuit seeking class-action status alleges the utility failed to adequately manage trees near its lines, contributing to repeated outages going back to August 2024, well before this storm. Indiana's governor ordered the state's utility consumer advocate to investigate not just the storm response but whether money regulators had already approved for reliability improvements was actually spent as promised, and some residents raised a separate question: whether NIPSCO's attention, and investment, had shifted toward serving new, power-hungry data centers instead of the lines already running through Gary's own neighborhoods.

Two weeks without power, in August heat, for the better part of a county, is exactly the scenario Chapter 4 is written for. But the deeper lesson isn't just about weather: a grid already flagged as fragile, serving a city with fewer resources to absorb a long outage, took the derecho's hit hardest and recovered slowest. Preparedness doesn't change who a utility invests in first; it changes whether a household is still waiting on that investment when the power finally goes out.

Do this: size your backup power for two weeks, not two days, Gary's own outage ran 14 days, and don't assume your neighborhood's grid will recover as fast as the one a few miles away. See Sizing a Solar Array & Battery System.

Water infrastructure failure & drought · Puerto Rico & the U.S. Virgin Islands · 2026

Documented September 2026

Puerto Rico and the U.S. Virgin Islands' Water Crisis

In mid-June 2026, engineers found a serious crack in the Superaqueduct, one of Puerto Rico's most important water transport systems. While repairs were underway, two additional failures turned up elsewhere in the same aging system. Combined, they cut off reliable running water to more than 100,000 people across San Juan, Bayamón, Guaynabo, Caguas, Juncos, Loíza, and other municipalities; as of mid-July, some residents had gone more than two months without dependable water. Puerto Rico's governor activated the National Guard to run tanker trucks through the hardest-hit neighborhoods, and households without another option reported spending $70 to $100 a day buying bottled or trucked water, a cost with no insurance category and no single event to file a claim against, repeated day after day. The underlying cause wasn't one storm; it was decades of deferred maintenance on a water system already stressed by drought, catching up all at once.

Forty miles east, the same pattern repeated on a different set of islands, for a different underlying reason. Since a Level 3 ("extreme") drought took hold across the U.S. Virgin Islands in mid-2026, households on St. Croix, St. Thomas, and St. John, many of which depend on rooftop-collection cisterns rather than a piped municipal supply, watched their private water reserves run down with no rain to refill them. At the tightest point, St. Croix had roughly three days of stored water left islandwide, St. John about four, and St. Thomas about four and a half. The drought exposed a second, longer-running problem: the territory's public water utility loses a staggering share of what it produces before it ever reaches a customer, line loss on St. Croix rose from 69% to 72.5% between 2024 and 2026, meaning nearly three gallons have to be treated and pumped for every one a household actually receives. Power failures on St. Thomas compounded the strain, since water production there depends on electricity.

Neither failure looked sudden from the outside, and neither one was, really: San Juan's system had been deferred for decades before it cracked, and the USVI's line losses had been climbing for years before a drought finally pushed reserves past what conservation could cover. Two U.S. territories, an ocean apart, learned the same lesson at the same time: aging infrastructure and a resource crisis don't fail independently, the second just exposes how little slack the first had left.

Do this: store more than the survival minimum and know a real backup source, households here paid $70-100 a day for trucked-in water with no insurance category and no single event to file a claim against. See Storing Water Correctly.

Drought · Corpus Christi, Texas · 2024–2026

Documented August 2026

Corpus Christi's Near Day Zero

By late April 2026, Corpus Christi's combined water reservoirs (Lake Corpus Christi and Choke Canyon) had fallen to just 7.8% of capacity after a prolonged drought, prompting mandatory Stage 3 water-use restrictions and warnings from local officials and water experts that the situation was, in their words, "tremendously grave." The city had already been under Stage 3 for well over a year by that point; the full run eventually reached 596 consecutive days before conditions improved.

What pulled Corpus Christi back from the edge was several things at once, not one fix: aggressive investment in alternative water supplies, conservation that cut average household use from roughly 6,000 to 4,500 gallons a month, and rain that happened to arrive in time to meaningfully refill the Nueces and Frio watersheds. By late summer 2026 the city had stepped down to Stage 2, and its next projected water emergency had been pushed back more than a year.

A major American city came within a stretch of bad luck, one more dry season, of an actual municipal water emergency, and the thing that ultimately saved it was largely outside anyone's control. That's a more unsettling lesson than a clean catastrophe: conservation and infrastructure investment bought time, but they didn't resolve the underlying shortfall on their own.

Do this: build your own real reserve rather than counting on rain arriving in time the way Corpus Christi did, restrictions can run for years, not weeks. See Storing Water Correctly.

Water scarcity · Utah & the Colorado River Basin · ongoing

Documented September 2026

The Great Salt Lake and the Colorado River's Record Lows

The slowest-moving entries on this list are two water systems, not one, and they're both running low at the same time. Both belong in Part One rather than Part Two's trend lines because both have already hit documented record lows; nothing below is a projection. The Great Salt Lake, the largest saltwater lake in the Western Hemisphere, hit its lowest recorded water level in October 2022 and remains critically low; the 2025 water year ended as the third-lowest on record. Driven by upstream water diversion and two decades of megadrought, Utah's own 30-year projection estimates it would take a sustained 800,000 acre-feet of additional inflow every year just to restore the lake to a healthy level by 2055. Exposed lakebed contains arsenic, lead, and other heavy metals; researchers studying the dust it kicks up have linked it to increased respiratory illness and cardiovascular disease for the more than 2.5 million people living along the Wasatch Front, effectively the entire population this manual is written for. Rising salinity is also collapsing the food web that supports migratory birds and the local economy tied to them.

A few hundred miles south, the same mechanism is playing out at a much larger scale. In August 2026, Lake Mead, the largest reservoir in the United States, fell to its lowest water level since Hoover Dam was built in the 1930s; Lake Powell was tracking toward breaking its own 2023 record the same month. Together, the two hadn't been this low since 1957, before Lake Powell existed as a reservoir at all. Roughly 40 million people across seven states rely on the Colorado River system these two reservoirs anchor, for drinking water, irrigation, and electricity; Lake Mead's level is now just above the point at which Hoover Dam's hydropower generation would be affected, and researchers project Lake Powell could reach "minimum power pool," unable to generate hydropower at all, by spring 2027. This isn't isolated to the two largest reservoirs either: Elephant Butte, New Mexico's largest, sat around 1.4% of capacity in mid-2026, Colorado's Blue Mesa around 24%.

This matters directly here, not just downstream: Lake Powell sits on the Utah–Arizona border, and Utah holds a legal water allocation from this exact system, on top of its own separate, already-strained supply covered by the Great Salt Lake above. Both are the same underlying story, a shrinking, earlier-melting mountain snowpack, not a lack of winter storms (the mechanism described in Part Two's Utah climate section), playing out at two different scales at once: single-lake for the Great Salt Lake, river-basin for the Colorado. Slow emergencies still end in real costs; they just give less warning that the bill is coming due.

Do this: treat long-term water security as a real project, not a one-time kit, this is exactly what Tier 3's self-sufficiency work exists for. See Legal Rainwater Harvesting.

Climate change · global, national & Utah

Updated September 2026

The same warming trend shows up at every scale

Global: the IPCC's 2023 AR6 Synthesis Report states plainly that human activity has "unequivocally" caused the warming measured since the pre-industrial era, with climate-related risks now assessed as higher than in the IPCC's own previous reports. A September 2026 UN Environment Programme report, "Limiting Overshoot," goes further: keeping long-term warming below 1.5°C is no longer possible. The world is already about 1.4°C warmer than pre-industrial levels, 2024 averaged 1.55°C, the warmest year on record, and at current emissions the remaining 1.5°C carbon budget will be exhausted within roughly three years, with the sustained threshold the Paris Agreement tracks now expected to be crossed before 2030. UNEP's own language: there are "no plausible scientific pathways" left to avoid it. Climate policy's focus is shifting accordingly, from preventing the breach to limiting how far past it warming rises.

National: NCA5 counted 89 billion-dollar weather and climate disasters between 2018 and 2022; NOAA's own data shows the average gap between one and the next shrinking from 82 days in the 1980s to just 12 days in 2023. A February 2026 Geophysical Research Letters study projects a greater than 90% chance U.S. disaster damages exceed $500 billion between 2026 and 2030, and a 54% chance they top $1 trillion. NOAA's 2024 tally: 27 such events, $182.7 billion in damage, a 14th consecutive year with 10 or more. NCA5 also projects the atmospheric rivers that deliver most of the West's winter precipitation growing stronger and wider, raising regional flood risk even as overall aridity increases.

Utah: the same trend shows up locally as a water problem, not a temperature headline. The state's annual mean temperature has risen about 0.4°F per decade over the last 40 years, with 3–10°F more projected by century's end. Winter precipitation totals haven't changed much, but rising temperatures have shifted more of it from snow to rain, cutting Utah's peak mountain snowpack 16% since 1979 and moving the date of peak snowpack about 9 days earlier. For a state with no major river system of its own, entirely dependent on that snowpack melting on schedule to fill reservoirs and the Great Salt Lake, a shrinking, earlier-melting snowpack is the mechanism behind the water crisis in Part One, not a separate risk.

That pattern produced a direct consequence in 2026: fueled by a historically dry summer, Utah burned more forest and woodland than in any year on record, past the previous 2018 record, with wildfire suppression spending projected to exceed $33 million, its highest ever. The Cottonwood Fire above Beaver, one of three fires this year to individually burn more than 60,000 acres, is the case study in Part One; this is the trend line behind it.

El Niño / ENSO · global & Utah

Updated September 2026

The Pacific is running the strongest El Niño in decades, layered on top of everything above

Every major forecasting body describes this event the same way: unusually strong, arriving fast. The World Meteorological Organization's August 2026 update shows multi-model consensus Niño 3.4 sea-surface anomalies exceeding 2.9°C, already beyond WMO's own threshold for a "strong" event. NOAA's Climate Prediction Center issued a formal El Niño Advisory in June 2026 and puts the probability of a historic event, exceeding every El Niño on record back to 1950, at 69% for October–December. The UK Met Office's own projections already show values exceeding 3°C, the most intense in over a century of record-keeping.

The weakened trade winds and shifted jet stream behind that warming reorganize weather worldwide, not evenly: NOAA's seasonal outlook shows the heaviest rain signal over coastal California and the Southeast, while a drier pattern holds elsewhere, Indonesia recorded more than 96,000 wildfire hotspots by late July 2026, more than at the same point in any year on record, as this El Niño, given a roughly 90% chance of becoming the strongest in about 150 years, intensifies the same dry-season pattern behind 2015's previous worst year. Damaging marine heat is projected to reach close to half the world's ocean surface by year's end, adding to a fourth global coral bleaching event that had already affected 84% of the world's reef areas by March 2025, the most intense such event ever recorded.

The economic and humanitarian toll follows the pattern of the comparably strong 1997–98 El Niño: the Peterson Institute projects a first-year global GDP loss of roughly $986 billion, about 0.8% of world output, growing to a six-year cumulative loss centered around $17.6 trillion. The World Food Programme projects 49 million additional people pushed into acute hunger by the end of 2027 across 45 already food-insecure countries. These impacts land hardest on lower- and middle-income tropical countries, but the mechanism that matters most here is closer to home: NOAA's winter outlook already favors below-normal precipitation for northern Utah this season, meaning the state's declining long-term snowpack trend could be compounded by an unusually strong El Niño winter arriving on top of it.

Seismic risk · Wasatch Front

Updated September 2026

The earthquake risk here is not hypothetical

In 2016, the Working Group on Utah Earthquake Probabilities (Utah Geological Survey and USGS) published the first formal forecast for the Wasatch Fault: a 43% chance of a magnitude 6.75+ quake, 57% for 6.0+, and 93% for 5.0+, all within the next 50 years. Roughly 85% of Utah's population, infrastructure, and economy sits within 15 miles of that fault. The 2020 Magna earthquake was, per the state's own scientists, ordinary background seismicity, a data point within the forecast, not an exception to it.

Probability is one question; consequence is another. In 2015, the Earthquake Engineering Research Institute's Utah Chapter, using FEMA's official HAZUS methodology, modeled a magnitude 7.0 earthquake on the Salt Lake City segment of the Wasatch Fault, the segment running beneath the most populated part of the state: 2,000 to 2,750 fatalities, 7,400 to 9,300 hospitalization-level injuries, and roughly 78,000 to 84,000 households displaced. It projects 55,400 buildings destroyed outright, 61% of all buildings across the 12-county region moderately damaged or worse, 444,000 homes losing power and 480,000 losing water. Short-term direct economic losses alone were estimated at $33.2 billion, a figure that, following the pattern already seen with Lahaina's initial estimate more than doubling within a year, almost certainly understates the true multi-year recovery cost.

The model attributes much of the projected toll to unreinforced masonry buildings, the same 90,000-plus structures discussed in the policy section below, whose retrofit funding depends on FEMA mitigation grants currently caught in litigation. This is what "the big one" is projected to actually do, not a hypothetical worst case, but the documented output of the same methodology FEMA uses to plan disaster response nationwide.

Infrastructure

Updated September 2026

The systems underneath are aging faster than they're being fixed

The American Society of Civil Engineers issues a national infrastructure report card every four years. In 2025, the U.S. earned an overall C, its best grade since ASCE started in 1998, but still a C. Energy infrastructure was downgraded to a D+, driven by concerns about grid capacity and safety. Roads improved only marginally, from a D to a D+. An aging grid and aging water systems are exactly the systems that failed in Winter Storm Uri, and exactly the systems every household preparedness plan has to assume might fail again.

Utah's own hazard planners have flagged the likeliest failure mode: the state's Enhanced State Hazard Mitigation Plan (FEMA-approved, 2024) states that in Utah, unlike the nation as a whole, winter storms cause more deaths than any other weather-related hazard. The grid Utah shares with the interconnected West has already been stress-tested for a worst case: the Western Electricity Coordinating Council's most recent reliability analysis modeled an extreme winter event causing up to 29 gigawatts of load loss over 16 hours across the Western Interconnection, and flagged growing risk of exactly that shortfall developing after 2026. No Utah-specific study has modeled a winter storm as precisely as the earthquake scenario above, but the outline is documented, and it points the same direction: aging systems, under mounting stress.

The equipment behind the grid is aging out as the supply chain to replace it collapses. More than 75% of the nation's distribution and large power transformers have already outlived their roughly 50-year design life. A replacement that took about 12 weeks to arrive before 2020 now takes an average of 128 weeks, more than two and a half years, and domestic manufacturers supply only about a fifth of national demand. Demand is rising into that same bottleneck: NERC's January 2026 assessment projects summer peak electricity demand growing 224 gigawatts over the next decade, 69% more than projected a year earlier, driven mostly by AI data centers, and names WECC's Basin and Northwest subregions, both covering Utah, among five regions nationally facing high reliability risk by 2030. The Gary, Indiana outage in Part One is what this looks like in practice, and worse: a grid old enough to need exactly this kind of investment, in a city where residents say the reliability spending regulators had already approved didn't show up on their own streets before the storm hit.

Fuel & refining

Updated September 2026

The system that supplies fuel is running with no spare capacity

Two large refineries left U.S. production this year, LyondellBasell's Houston refinery (264,000 barrels a day, early 2025) and Phillips 66's Los Angeles-area refinery (139,000 barrels a day, end of 2025), with no announced timeline for replacing either. What's still running is doing so at close to full tilt: the EIA reported national refinery utilization at 97.2% in mid-August 2026, near the highest level in the agency's records back to 1990, leaving almost no spare capacity to absorb an unplanned outage. EIA's February 2026 outlook projected the country's combined fuel stockpile would end the year at 375 million barrels, the lowest since 2000.

California shows what a fully strained system looks like in practice. The state already imports roughly 40% of its gasoline and 35% of its jet fuel and diesel; Valero's Benicia refinery near San Francisco was slated to close in April 2026, and further closures among the state's remaining refineries would push those import shares well past 50%, into ports and pipelines that were never built to carry that volume.

The exposure to an external shock isn't hypothetical: the Strait of Hormuz closure in Part One pushed the national average diesel price to $5.37 a gallon within weeks, hitting a refining and inventory system with little room to absorb it. The pipelines moving fuel domestically carry their own aging-infrastructure risk: 43% of crude and refined-product pipelines, and 54% of natural gas transmission pipelines, were installed before 1970, and corrosion's share of reported pipeline incidents rose from about 18% to more than 25% in 2024.

Trucking & freight

Updated September 2026

Moving goods costs more, with a shrinking fleet to do it

The American Transportation Research Institute's 2025 survey put the average cost of operating a truck at $2.336 a mile, the highest on record and up 3.4% from 2024. Freight rates haven't kept pace: truckload carriers averaged an operating margin of just 0.4% in 2025, the industry's third consecutive year of a freight-market recession. Carriers responded by cutting truck counts, the largest reduction in industry capacity since the recession began in 2022, trimming exactly the fleet that would be needed to absorb a sudden surge in demand, whether from a disaster, a supply-chain shock, or an emergency response effort.

Roads & bridges

Updated September 2026

The network is aging faster than the "fair" category can absorb

Roads earned a D+ in ASCE's 2025 report card, bridges held at a C. The bridge data tells the more precise story: 49.1% of the nation's roughly 623,000 bridges are now "fair," up from 43.3% in 2012, while only 44.1% remain "good," meaning the fair category, the one that precedes "poor," is growing faster than bridges are being renewed. Just under 7%, about 46,000 bridges, are already structurally deficient, and the total rehabilitation backlog stands at $191 billion. Federal investment under the 2021 infrastructure law has closed part of the gap, but ASCE flags that funding beyond 2026 is uncertain, right as the backlog would need two decades of sustained spending to meaningfully shrink.

Water infrastructure

Updated September 2026

The pipes are older than the funding to replace them

Water infrastructure earned its own dedicated grades in ASCE's 2025 report card, none improved from 2021: drinking water sits at a C-, wastewater at a D+, stormwater at a D. The gap is largely financial. The American Water Works Association's 2026 assessment puts the true cost of sustaining the nation's drinking water systems through 2050 at $2.1 to $2.4 trillion, against current spending of roughly $33.6 billion a year toward a needed $90.2 billion, a $56.6 billion annual shortfall. Wastewater and stormwater face a similar squeeze, about $99 billion a year in need with only around 30% funded, a gap the Value of Water Campaign projects will grow past $690 billion by 2044.

Lead service lines are the sharpest edge of that gap, and even the basic count is contested: the EPA's 2024 survey put the national total at more than 9 million lead service lines still in use; in November 2025 the agency cut that estimate by more than half, to about 4 million, a revision environmental groups have urged the public to treat with skepticism given the stakes of undercounting a contaminant with no safe exposure threshold. Either figure describes the same problem visible in Part One's Flint case study: millions of households still drink water through pipes regulators say a well-funded system should have replaced years ago.

Cybersecurity

Updated September 2026

Attacks on infrastructure itself are rising

This is the newest category of risk here, and it's accelerating. Ransomware groups have shifted from opportunistic targets to deliberately targeting critical infrastructure: the FBI and CISA have flagged Akira (over $244 million in claimed proceeds as of late 2025) and RansomHub (which has hit water, wastewater, and government-services targets specifically) as active threats to the systems that keep the lights and water on. Colonial Pipeline in 2021 showed what a single successful attack can do; the current trend is more attempts, against more of exactly that kind of target.

Agriculture has become a specific target within that same trend: the Food and Agriculture Information Sharing and Analysis Center tracked 72 distinct active threat actors targeting the food supply chain in its 2025-2026 reporting, split between financially motivated ransomware crews (a sector-wide 82% surge in 2025) and nation-state actors. Separately, a 2014 FERC analysis, still cited by grid-security researchers today, concluded that a coordinated physical attack knocking out just nine of the country's roughly 55,000 transmission substations, on a high-demand day, could collapse the entire U.S. grid for at least 18 months, largely because so few domestic factories build the transformers needed to rebuild it, the same manufacturing bottleneck described above.

Cargo theft has followed the same cyber-enabled pattern: the FBI reported 2026 cargo theft losses across the U.S. and Canada reached nearly $725 million in 2025, a 60% jump from 2024, as criminal groups compromise freight brokers' computer systems to hijack shipments and flood load boards with fraudulent listings.

Food supply & safety

Updated September 2026

The system that grows, moves, and inspects our food is strained on every front

Farm finances are deteriorating on their own trend line, independent of any single disaster. Chapter 12 farm bankruptcy filings rose 46% in 2025, the third consecutive year of increases, as weak crop prices, softening livestock margins, and elevated interest rates squeezed producers already carrying more debt. USDA projects total farm debt will reach a record $624.7 billion in 2026, with interest expense alone hitting a record $33 billion.

The system's exposure to federal disruption is real and recent: the 43-day federal government shutdown (Oct. 1–Nov. 12, 2025), the longest in U.S. history, furloughed roughly 95% of the Natural Resources Conservation Service's staff and delayed farm loans and disaster assistance during fall harvest, on top of roughly 20,000 USDA staff departures since the start of 2025.

The system meant to catch a contaminated-food outbreak before it spreads has itself been cut back. On July 1, 2025, the CDC reduced its 30-year-old FoodNet surveillance network from tracking eight foodborne pathogens to two, Salmonella and E. coli, making Campylobacter, Cyclospora, Listeria, Shigella, Vibrio, and Yersinia optional to report, citing funding that "has not kept pace" with costs. Ten months later, an outbreak of cyclosporiasis traced to Taylor Farms iceberg lettuce became the largest on record for that parasite, sickening at least 10,930 people across 17 states, hospitalizing 454, and killing two, months after the pathogen behind it had been made optional to track.

The agencies behind that surveillance have lost the people who ran it. FDA shed 3,859 employees in 2025 and another 473 in 2026; its food inspection workforce is now about 20% understaffed. The GAO found FDA is averaging just 917 foreign food facility inspections a year against a Food Safety Modernization Act target of 19,200, and isn't inspecting 49% of high-risk domestic facilities or 95% of the foreign facilities it's responsible for. Congress separately pushed back the Food Traceability Rule, the regulation meant to trace a contaminated product to its source quickly, by 30 months, to July 2028.

Funding is moving the same direction: the FY2026 budget request cuts FDA funding 3.9% from 2025's level, on top of the workforce already lost. None of this shows up yet as a cleaner recall system: food recalls totaled 57.4 million units in the first quarter of 2026 alone.

Medical access

Updated September 2026

Getting care is getting harder in most of the country

More than 120 million Americans, over 80% of U.S. counties, live in what GoodRx's 2025 research classifies as a healthcare desert: inadequate access to a pharmacy, primary care, a hospital, or emergency care. Pharmacy deserts have expanded since 2021, from 41 million to more than 48 million people affected, as more than 1,300 pharmacies have closed nationally; 28 million people now live more than 30 minutes from a hospital.

Coverage is trending the same direction: the CBO projects uninsured Americans growing from 30 million in 2026 to 37 million by 2036, driven mainly by Medicaid/CHIP reductions in the 2025 federal reconciliation act and the expiration of enhanced ACA subsidies. Medicaid enrollment has already fallen roughly 20% nationally from its pandemic-era peak.

Workforce projections vary by methodology, from AAMC's estimate of up to 86,000 too few physicians by 2036 to HRSA's estimate of more than 124,000 too few by 2027, but every projection points the same direction: fewer doctors relative to a growing, aging population, on top of the access gaps above.

Disease & pandemic preparedness

Updated September 2026

The systems built after 2020 are being run down, not built up

Trust for America's Health's 2026 "Ready or Not" report, an annual assessment of state and federal public health emergency readiness, finds preparedness worsening rather than improving: 20 states scored in the top tier, 17 and D.C. in the middle, 13 in the lowest, and the report flags HHS's March 2025 termination of more than $12 billion in COVID-era grants that states had broadened to cover infectious-disease surveillance and general emergency preparedness. The Pandemic and All-Hazards Preparedness Act, the core federal law authorizing much of this readiness infrastructure, has been running on temporary extensions since its authorization lapsed at the end of 2023, and remains unreauthorized.

The threats this infrastructure exists to catch haven't paused. The U.S. recorded more than 5,000 confirmed measles cases between January 2025 and August 2026, the worst year since the early 1990s and a threat to the nation's measles elimination status; Pennsylvania reported the state's first measles deaths in 35 years in August 2026, both in unvaccinated people. H5N1 avian influenza has infected at least 71 people in the U.S. since 2024, mostly through infected dairy cattle and poultry (CDC still assesses general public risk as low). Florida logged 23 locally acquired dengue fever cases by mid-August 2026, a mosquito-borne disease increasingly gaining a foothold rather than arriving only via travelers.

Population & exposure

Updated August 2026

More people are living in the path of these risks

Utah was the fastest-growing state in the nation for most of the last decade, and growth remains concentrated exactly where the seismic risk is highest: the four Wasatch Front counties (Salt Lake, Utah, Davis, and Weber) accounted for nearly 73% of the state's total population growth between 2023 and 2024. The Kem C. Gardner Policy Institute projects Utah's population will grow from roughly 3.6 million today to 5.6 million by 2065. None of that growth changes the fault's probability of rupturing; it changes how many people and how much infrastructure are sitting on top of it when it does.

Insurance markets

Updated August 2026

The insurance industry is quietly agreeing with all of this

Insurance companies price risk for a living, and their behavior over the last several years is itself a data point. In California, one in five homes in the highest wildfire-risk areas has lost private insurance coverage since 2019; the state's FAIR Plan grew from covering 210,000 homes in 2020 to more than 463,000 in 2024, more than double what its total liability was two years earlier. Texas, Oregon, and parts of Arizona are following a similar trajectory. When the industry whose entire business model depends on correctly pricing disaster risk responds by leaving markets or radically raising prices, that is a real signal, not a talking point.

Policy & institutional capacity

Updated September 2026

The agencies built to warn and respond are themselves being resized

Everything above this point has been a hazard or a structural condition: weather, geology, aging pipes. This one is different: it's a set of active policy and budget decisions, made by the current federal administration and a Trump-appointed review council, that reduce the capacity of the specific agencies this document has already cited for basic facts throughout Parts One and Two.

In February 2025, NOAA terminated roughly 880 employees, including National Weather Service meteorologists, radar specialists, and Hurricane Hunter flight crew, as part of the Department of Government Efficiency's push to shrink the federal workforce. By mid-2025, about 40% of the nation's 122 weather forecast offices had significant staffing vacancies, and at least eight had cut or were preparing to cut overnight forecasting entirely. Congress later restored most of NOAA's funding and the agency began a partial rehire, but as of mid-2026 the National Weather Service remains hundreds of positions short. This is the same warning system this document opened with in the Lahaina case study: sirens and alerts are only as good as the people staffing them.

FEMA has seen two separate changes. In April 2025, the agency canceled its Building Resilient Infrastructure and Communities (BRIC) program, which had awarded more than $5 billion nationally and that Utah communities rely on directly: Salt Lake City's "Fix the Bricks" program, which retrofits the unreinforced masonry buildings discussed in the seismic-risk section above, applies for exactly this kind of funding every year. Twenty states sued, and in December 2025 a federal judge ruled the cancellation unlawful and ordered the program reinstated; FEMA said it would comply but gave no timeline. Separately, a FEMA Review Council created by executive order in January 2025 recommended in May 2026 that FEMA be restructured into a smaller, "less Washington-centered" agency, with more disaster cost and responsibility shifted onto state and local governments.

CISA, the agency that tracks the ransomware activity described earlier in this section, is losing staff on a similar trajectory: workforce fell from about 3,400 employees at the start of fiscal year 2025 to roughly 2,400 by December, through layoffs, buyouts, and early retirements, and the administration's fiscal year 2026 budget request would cut roughly 1,000 more positions. Senate appropriators have pushed back and proposed funding closer to 2025 levels, so the final number isn't settled.

None of this is fully resolved: courts and Congress have already reversed or softened several of these cuts. But the direction is consistent across four systems this document depends on for basic facts: less forecasting and cyber-defense staffing, less mitigation funding, and a push to move disaster-response cost onto the same state, local, and household budgets Part One showed can't easily absorb it.

A related but distinct problem sits underneath the staffing cuts above: the data itself. The Environmental Data and Governance Initiative has documented more than 500 changes to federal climate and environmental web content since January 2025, including the removal of NCA5 (the same report this document cites throughout Part Two) from federal websites as of June 2025 with no public explanation, the redirection of climate.gov to a narrower NOAA subpage, and EPA language changes replacing "climate change" with terms like "climate variability." This document cites federal sources throughout because they are, for now, still the most authoritative data available, and it's why every .gov citation in the Sources section below is paired with an archived snapshot: a live government page can move or disappear without notice in a way an independent, timestamped archive can't. Several universities and former federal scientists have taken the same approach at larger scale, mirroring NCA5 and other removed datasets on independent infrastructure.