Weather folklore persists because it is memorable and often sounds physically reasonable. What makes the five myths here worth correcting specifically is that each one does not just fail to help; it actively points toward a decision that is measurably more dangerous than the correct one.
01 — Lightning never strikes the same place twice
Tall, isolated structures get struck repeatedly, and the physics explains why
The myth
"Lightning never strikes the same place twice."
The National Weather Service directly refutes this on its official lightning myths page: lightning often strikes the same place repeatedly, especially a tall, pointy, isolated object. The Empire State Building is struck by lightning roughly 23 to 25 times per year, according to NOAA, and has been struck as many as a dozen times during a single storm. Lightning strikes wherever the physical conditions favor it in that moment, height, shape, and isolation from other tall objects, not based on any memory of where it struck previously. This is exactly why tall structures, radio towers, mountain peaks, and isolated trees, are routinely fitted with dedicated lightning protection systems rather than assumed to be safe after an initial strike.
The practical danger in this myth is what it implies about safety, not just its inaccuracy. Someone who believes a recently struck area is now "used up" and therefore safer may take more risk in that location during a subsequent storm, exactly the opposite of what the physics supports. NWS guidance remains the same regardless of strike history: if you can hear thunder, you are within lightning's striking distance, and the only reliable protection is a substantial enclosed building or a hard-topped vehicle, not an assumption about probability based on past strikes.
- The Empire State Building is struck roughly 23 to 25 times per year. Tall, isolated, pointed structures are struck repeatedly because height, shape, and isolation, not history, determine where lightning goes.
- Lightning protection systems on skyscrapers exist because repeat strikes are expected, not rare. Engineers design for recurring strikes on tall structures, not one-time events.
- Believing this myth can lead to riskier behavior, not safer behavior. Treating a previously struck area as "safer now" contradicts the actual physics of lightning formation.
- The safety rule does not change based on strike history. If you can hear thunder, seek shelter in a substantial building or hard-topped vehicle, regardless of whether the area has been struck before.
02 — A highway overpass is a safe tornado shelter
A 1991 video created this myth; a 1999 tornado outbreak proved it fatal
The myth
"A highway overpass is a safe place to shelter from a tornado."
This myth traces to a specific, widely broadcast piece of video: during a 1991 tornado near Andover, Kansas, a television news crew took shelter under a Kansas Turnpike overpass and survived with only minor injuries, and the footage was shown across the country. What that footage did not convey was how unusual the circumstances were, a relatively weak tornado, an overpass of unusual design, and a storm track that shifted at the last moment and did not directly strike the overpass. The myth persisted until the May 3, 1999 Oklahoma and Kansas tornado outbreak, when three separate highway overpasses were directly struck by tornadoes; each of the three produced at least one fatality among the people sheltering there, a result the NWS documented directly in its own post-event analysis.
The physical reason an overpass fails as shelter is specific: the embankment beneath an overpass sits higher than the surrounding terrain, and wind speed increases with height, so people sheltering there are already exposed to stronger wind than someone at ground level nearby. The confined space under the span can also create a wind-tunnel effect that further accelerates both wind speed and flying debris. Most overpasses are also fully open underneath, offering no protective girders or enclosed crawlspace, which leaves anyone there directly exposed to debris traveling at damaging speeds even in a comparatively weak tornado.
- The myth's origin, a 1991 news video, involved highly unusual circumstances. A weak tornado, an atypical overpass design, and a storm that shifted away from the bridge at the last moment all contributed to that crew's survival; none of these can be counted on.
- The 1999 Oklahoma/Kansas outbreak directly disproved the myth. Three separate overpasses were struck; each produced at least one fatality, an outcome NWS documented in its own official analysis.
- An overpass's elevation and shape actively increase wind exposure. Height increases wind speed, and the confined space under a span can create a wind-tunnel effect that accelerates both wind and debris further.
- If driving and no sturdy building is available, get out of the vehicle and lie flat in a low ditch, away from the vehicle and any overpass. This is the NWS-recommended fallback when no substantial shelter can be reached in time.
03 — Hurricanes only affect coastal areas
Inland flooding now kills more people than storm surge, sometimes hundreds of miles from the coast
The myth
"Hurricane damage is limited to coastal areas near where the storm makes landfall."
National Hurricane Center fatality research directly contradicts this. Analysis of the most recent decade of U.S. tropical cyclone deaths found that freshwater, inland flooding from rainfall was responsible for roughly 57 percent of direct fatalities, compared with about 11 percent from storm surge specifically, a striking reversal from decades earlier when storm surge was the leading killer. A separate peer-reviewed analysis found that roughly 80 percent of direct U.S. hurricane fatalities since 1970 occurred outside the county where the storm made landfall, with most of those deaths caused by inland flooding rather than wind or surge. A hurricane's rainfall is not tied to its wind category either; some of the heaviest rainfall totals on record come from storms that weakened after landfall but stalled or moved slowly, dumping rain over the same inland area for days.
Hurricane Floyd in 1999 illustrates the pattern directly: of the 56 people who died, 50 drowned in inland flooding, not storm surge. Beyond flooding, hurricanes routinely spawn tornadoes and cause extended power outages and wind damage hundreds of miles from the coast, well beyond any reasonable definition of "coastal impact." The practical danger in this myth is that it can lead inland residents to treat hurricane warnings as irrelevant to them specifically, when the National Hurricane Center's own data shows inland flooding, not the storm's landfall point, is now the leading cause of hurricane-related death in the United States.
- Inland freshwater flooding causes roughly 57% of recent direct hurricane deaths, versus about 11% from storm surge. This reverses the historical pattern and reflects real, current National Hurricane Center fatality data.
- Roughly 80% of direct U.S. hurricane deaths since 1970 occurred outside the landfall county. Most were from inland flooding, not wind or surge, and some occurred hundreds of miles from the coast.
- Rainfall intensity is not tied to wind category. A weakened or slow-moving storm can produce catastrophic inland flooding regardless of its category at landfall.
- Tornadoes, power outages, and wind damage from a hurricane extend well inland too. Treat hurricane warnings as relevant based on your actual forecast exposure, not your distance from the coastline.
04 — It can be too cold to snow
Cold air holds less moisture, but it can still snow at any temperature below freezing
The myth
"It can be too cold outside for snow to form."
NOAA's National Severe Storms Laboratory is direct about this: snow can occur at any temperature below 32°F, and there is genuinely no such thing as "too cold to snow," only conditions that make heavy snowfall less likely. The confusion has a real physical basis. Colder air holds less water vapor than warmer air, so as surface temperatures drop, the atmosphere generally has less moisture available to produce snow, and the heaviest snowfalls typically occur when the air near the ground is relatively close to freezing rather than deeply cold. But "less moisture available" is not the same claim as "no snow possible." Even extremely cold, dry air can still produce light snow or fine ice crystals, and importantly, the air aloft at higher altitudes, where snow actually forms, can carry meaningfully more moisture than the frigid air at the surface.
The National Snow and Ice Data Center notes the more accurate version of this idea is that it can be too warm to snow, since precipitation simply falls as rain once temperatures rise enough, but it cannot be too cold for snow to form, only too dry. The Dry Valleys of Antarctica illustrate this precisely: they are extremely cold but also extremely dry, with strong winds that strip away what little moisture is present, and as a result receive very little snow despite the extreme cold, a scarcity driven by dryness, not temperature.
- Snow can occur at any temperature below 32°F. There is no temperature floor below which snow physically cannot form; only moisture availability limits it.
- Colder air holds less moisture, which is why heavy snowfall is less common at very low temperatures. This is a real, correct physical fact, but it explains reduced snowfall amounts, not an absence of snow entirely.
- Air aloft, where snow forms, can carry more moisture than frigid surface air. Warm air higher in the atmosphere can supply the moisture needed even when the ground-level temperature is very cold.
- It can be too warm to snow (precipitation falls as rain instead), but never too cold. Extremely cold, extremely dry regions like Antarctica's Dry Valleys receive little snow because of dryness, not because of temperature alone.
05 — Taping windows protects them during a hurricane
Tape does not strengthen glass, and it can make the resulting shards more dangerous
The myth
"Taping windows in an X pattern helps protect them from breaking during a hurricane."
The Federal Alliance for Safe Homes (FLASH), a nonprofit that partners directly with FEMA, the National Weather Service, and state emergency management agencies on hurricane safety messaging, has run a dedicated public campaign called "Go Tapeless" specifically to correct this myth, after surveys found nearly 70 percent of homeowners still believed taping was effective. Tape adds no structural reinforcement to glass. A window fails under wind pressure and flying debris because of the force applied to the pane, not because of a lack of adhesive holding it together, and standard residential glass can begin to fail at wind speeds well below Category 1 hurricane strength.
The specific danger tape introduces is worse than doing nothing. Untaped glass typically fractures along natural stress lines into many smaller, relatively blunt pieces. Tape can interrupt that fracture pattern, holding fragments together momentarily and causing the pane to release as fewer, larger, more sharply edged shards once the wind pressure finally breaks the bond, shards capable of causing significantly more severe lacerating injuries than the smaller fragments an untaped window produces. FLASH, FEMA, and the National Hurricane Center are unified on the correct alternatives: properly installed hurricane shutters, impact-rated windows, or, as a last resort, plywood cut to fit and mechanically fastened into the structural frame, not simply nailed to exterior stucco or trim.
- Tape adds no structural strength to glass. Windows fail from wind pressure and debris impact, a force problem tape does nothing to address.
- Tape can produce larger, more dangerous glass shards than an untaped break. By holding fragments together briefly before releasing, tape can cause fewer, larger, sharper pieces to break free at once.
- FEMA, NWS, and FLASH all explicitly recommend against window taping. The "Go Tapeless" public awareness campaign exists specifically because the myth remains widely believed despite official correction.
- Use real protection instead: shutters, impact-rated glass, or properly mounted plywood. Plywood should be mechanically fastened into the structural frame, not simply nailed into stucco or trim, to actually resist wind pressure.
Quick reference
- Lightning strikes the same tall, isolated structures repeatedly. The Empire State Building is hit about 23 to 25 times a year.
- Highway overpasses are dangerous tornado shelters, not safe ones. Elevation and the confined shape increase wind exposure; three overpasses struck in 1999 each produced a fatality.
- Inland flooding, not storm surge, is now the leading cause of hurricane deaths (about 57% vs. 11% in recent data), and 80% of hurricane deaths since 1970 occurred outside the landfall county.
- Snow can form at any temperature below 32°F. Extreme cold reduces available moisture and heavy snowfall, but it never makes snow impossible; only extreme dryness does that.
- Taping windows does not strengthen glass and can create larger, more dangerous shards. Use hurricane shutters, impact glass, or properly fastened plywood instead.
Primary sources
- National Weather Service: Lightning Myths and Facts: the direct refutation of the "never strikes twice" myth and the Empire State Building strike-frequency figure.
- NOAA: 5 Striking Facts Versus Myths About Lightning: confirmation of the Empire State Building's approximate 25-strikes-per-year figure.
- National Weather Service Mobile, AL: Severe Weather Awareness, Tornado Safety: the explicit guidance against sheltering under an overpass, and the low-ditch fallback for drivers.
- Overpass myth (citing NWS analysis of the 1991 Andover, KS and 1999 Oklahoma/Kansas events): the origin of the myth and the three overpass fatalities during the May 3, 1999 tornado outbreak.
- American Meteorological Society: Recent Trends in Tropical Cyclone Fatalities in the United States: the 57% freshwater-flooding versus 11% storm-surge fatality breakdown for the most recent decade studied.
- An Analysis of Coastal and Inland Fatalities in Landfalling US Hurricanes, Natural Hazards journal: the finding that roughly 80% of direct U.S. hurricane fatalities since 1970 occurred outside the landfall county.
- NOAA National Severe Storms Laboratory: Severe Weather 101, Winter Weather FAQ: the explanation that "too cold to snow" really reflects insufficient lift and moisture, not a temperature floor on snow formation.
- National Snow and Ice Data Center: Science of Snow: confirmation that it can be too warm, but never too cold, for snow to form, and the Antarctic Dry Valleys example.
- Federal Alliance for Safe Homes (FLASH): Taping Windows to Prep for Hurricane Is Dangerous: the "Go Tapeless" campaign, the FEMA/NHC partnership on this messaging, and the survey finding nearly 70% of homeowners still believed the myth.