Hourwiz Explains · video notes

What If the Sun Disappeared? Our full timeline model

From the Hourwiz video “The Sun is gone.” · Checked Oct 16, 2026

In the video we switched off the Sun in code and ran the clock forward. Here's the whole model: how it works, every number it gave, where it agrees with published estimates, what it leaves out, and the files, so you can check our work.

First: it can't really happen

The Sun can't just vanish. A physics professor at West Texas A&M puts it plainly: “the sun can't really suddenly disappear.” NASA says the Sun “will last another 5 billion years or so before it becomes a white dwarf.” So this is a thought experiment: we switch off the Sun's light and its pull at the same moment and watch what spreads outward.

Why it matters: the Sun holds 99.8% of the solar system's mass, and its gravity holds the solar system together (NASA).

Our method

Our model. One Node script we wrote, no outside libraries, run on Oct 16, 2026 with Node v24.21.0. Every input is a published value listed in the sources below. We set the vanish moment at Thu Oct 29, 2026, 5 PM EDT; the rounded results on this page hold on any date.

• (a) Light and gravity delay. Each change travels at the speed of light: the time is distance ÷ c, from JPL's constants. For the Moon and planets we add the extra path their reflected light takes, for 4,383 vanish dates (12 years).
• (b) Orbit sim. Earth and the Moon, stepped with the RK4 method, starting from JPL's planet positions for the date. Each body loses the Sun's pull only when the change reaches it (its distance ÷ c), then flies free. Mars, Jupiter and Saturn switch to straight lines on their own clocks. We check the Moon from 4 starting positions over 10 years, and test whether Jupiter could ever capture Earth, on all 4,383 dates.
• (c) Cooling model. A land column (a surface air layer on 20 km of rock, with Earth's own heat flowing up from below) and an ocean column (the top 50–100 m of sea cools to its freezing point, then an ice lid grows). Three settings span an honest range: a warm edge (today's greenhouse kept, the whole air column, rock conductivity 3.0, a 100 m sea layer), a cold edge (no greenhouse, the lowest tenth of the air, 2.0, 50 m) and a middle case in between.
• (d) The 1-second blink. The Sun's pull off for exactly 1 second, then back. We compare a test Earth with its twin for 2 years.

The formulas in our script

delay = distance ÷ c
  1 au: 499.0 s = 8 min 19.0 s

ghost light = (Sun→planet
  + planet→Earth − Sun→Earth) ÷ c

after the pull: a straight line,
  same speed (no force, no change)

blink drift = ½ × (GM_sun ÷ r²) × (1 s)²
  about 3 mm at 1 au

heat out = ε × σ × T⁴
  ε = 0.605 (today's greenhouse)
      to 1.0 (none)

ice conductivity k(T) = 651 ÷ T
  (W/m/K, T in kelvin)

floor = (0.087 W/m² ÷ σ)^¼
  = 35.2 K = −396 °F

How we checked the model

• Light time for 1 au: 499.004784 s, matching JPL's 499.004783836 s.
• Earth's distance (148.585 million km) and speed (29.987 km/s) on the date fall inside NASA's closest-to-farthest and slowest-to-fastest ranges.
• After the pull is gone, Earth's speed stays constant for 10 years, and the 1-day distance equals speed × 1 day.
• The Moon stays with Earth from all 4 starting positions (384,397–384,403 km apart over 10 years).
• Earth and Jupiter always pass too fast to pair up, on all 4,383 vanish dates.
• The 1-second blink: 3.003 mm in the sim vs 3.006 mm from the textbook formula (within 2%).
• Every land curve cools steadily and stays above its floor; no ice grows thicker than its balance point.

The script printed ALL CHECKS PASS. Where the run disagreed with our hand estimates, we used the run.

The timeline at a glance

Our model, sim and math. Times count from the moment the Sun vanishes. Ranges cover our warm edge to cold edge, or every vanish date we tried.

8 minutes 19 seconds of nothing

• Sunlight is about 8 min 19 s old when it reaches us at Earth's average distance (JPL: 499.0 seconds per astronomical unit). Across the year it ranges from about 8 min 11 s at our closest to the Sun to about 8 min 27 s at our farthest (our calc from NASA's distances). On our vanish date: 8 min 15.6 s, at 92.33 million miles.
• Changes in gravity travel at the speed of light. In 2017, the LIGO and Virgo detectors caught ripples in gravity from two neutron stars colliding about 130 million light-years away, and NASA's Fermi space telescope caught a flash of gamma rays about 2 seconds later. The paper puts the difference between the speed of gravity and the speed of light at between −3 × 10⁻¹⁵ and +7 × 10⁻¹⁶ of the speed of light.
• Our math on that race: a 130-million-year trip, finished 1.74 seconds apart, is about 4 × 10⁻¹⁶ of the trip. If gravity were even 1% slower than light, it would have shown up about 1.3 million years late.
• So Earth keeps circling the empty spot for those 8 minutes, and leaves its orbit at the same moment the sky goes dark. The old idea that gravity is instant, so Earth flies off right away, is out of date.
• No sunset, no twilight. The blue sky is just scattered sunlight (NASA), so the day side goes dark at once: the extra path through about 100 km of air adds only 0.33 thousandths of a second (our calc).

Where Earth goes

Our sim · JPL data. Vanish moment Oct 29, 2026. The rounded wording holds on any date.

• The Moon comes with us. In our sim it stayed with Earth from all 4 starting positions, 384,397–384,403 km away for 10 years, circling every 27.28 days. To escape it would need about 41% more speed: 1.449 km/s against its 1.025.
• Each planet leaves only when the change reaches it: Jupiter after about 43 minutes and Saturn after about 80, at their average distances (44 min 12 s and 78 min 24 s on our date).
• Does Jupiter catch us? No. Jupiter flies off too. On our date the closest the two get is 3.50 au (325 million miles), after 222 days, passing at 33.4 km/s; Jupiter's escape speed there is 0.70 km/s. Over all 4,383 dates, the closest pass is 0.014 au (1.3 million miles) at 23.9 km/s: a flyby, not a capture.
• Tides keep going, just smaller: the Moon does most of the work, and the Sun's part is about half the Moon's (NOAA).
• Earth becomes a planet drifting between the stars. Scientists call these “rogue planets”.

Would it be pitch black?

• The stars stay. They're in the sky in daytime too, hidden by scattered sunlight (StarDate, University of Texas McDonald Observatory). They're other suns, so they don't care that ours is gone.
• The Moon goes dark a second or two after our sky. It has no light of its own (NASA); the reflected light from a full Moon takes about 2.6 s longer to reach us (our calc).
• Our own lights stay on, as long as the power does.

The extra time is the longer path the light takes: Sun to planet, then planet to Earth, minus Sun to Earth. Uranus and Neptune are left out: most people can't see them without a telescope.

How cold would Earth get?

Our model, shown as a band. Far inland means no ocean heat blown in. The model has no winds or currents, so real coasts would sit between our land and ocean curves.

• The sea barely notices at first: the sea surface is still about 57–58 °F after a week, 50–56 °F after a month and 34–51 °F after 3 months in our model.
• A crude average for the whole planet (71% ocean, 29% land): about 28–53 °F after a week, and about −157 to −12 °F after a year.
• The record we compare with: −89.2 °C (−128.6 °F), Vostok station, Antarctica, July 21, 1983 (WMO).
• The air freezes out in the end: oxygen and nitrogen turn liquid and fall to the ground, after centuries to thousands of years in our rough model. It doesn't model the heat the air gives off as it turns liquid, so the real dates would be later. The air doesn't float away: gravity doesn't change.
• The floor: about −396 °F (35.2 K, −238 °C), where the heat leaking out of Earth's inside (0.087 W/m², a peer-reviewed average) balances what the surface radiates to space (our math).

Our model vs published estimates

• Popular Science (2008) estimates the average surface temperature would drop below 0 °F within a week, and dip to −100° within a year. Our middle case agrees far inland (0 °F at 7.4 days), but not for the whole planet: in our model the oceans hold their heat for months, and the crude global average after a week is still 28–53 °F. The −100° after a year sits inside our 1-year planet band.
• The long run: in the same article, Caltech planetary scientist David Stevenson puts it at a stable −400° after millions of years. Our floor of −396 °F agrees.

Would the oceans freeze?

Seawater freezes at about 28.4 °F (−2 °C), because of the salt in it (NOAA).

• The ice works like a blanket. One published estimate (Popular Science) says it would keep the deep water from freezing solid for hundreds of thousands of years.
• But not forever. A peer-reviewed study (Abbot & Switzer, 2011) finds that a planet would need about 3.5 times Earth's mass for its own heat to keep a liquid ocean under the ice. Our model agrees: the ice stops thickening only at 14,338–15,278 m (about 9 miles), deeper than any ocean.
• What our model leaves out: salt, the heat stored in the deep water, and hot spots like vents. So the real dates are probably later. It's a 1-D model: one column of ocean, no currents.

What could still live?

• Photosynthesis stops the moment the light does (Popular Science; NOAA lists light as one of its inputs).
• Life at deep-sea vents doesn't run on sunlight. Bacteria and archaea turn chemicals from the vents into food, and they support hundreds of species of animals (NOAA). Honest footnote: many vent bacteria still use oxygen dissolved in seawater (NOAA).
• Microbes live deep inside rock, on energy from rocks. Together they hold 15 to 23 billion tonnes of carbon: 245 to 385 times the carbon in all humans (Deep Carbon Observatory, 2018).
• Heat from underground already works. About 9 in 10 homes in Iceland are heated with geothermal heat (Samorka, Iceland's energy and utilities association), and Iceland grows tomatoes through the dark winter in greenhouses heated by hot springs, under electric lights. Popular Science picked Iceland as “one good place to camp out”.
• How long would people survive? We didn't find a sourced number, so we don't give one.

Bonus: if the Sun vanished for just 1 second

• We'd get 1 second of darkness, about 8 min 19 s later: the missing light and the missing pull both arrive one light-time late.
• During that second, Earth's path bends by about 3 mm, about ⅛ inch (our sim: 3.003 mm; formula: 3.006 mm).
• Afterwards Earth's orbit wobbles about 30 km (20 mi) in and out and falls up to about 116 km (72 mi) behind, about 4 seconds late, at half a year. A year later it's back to within about 9 km (a third of a second). It's a repeating wobble that never builds up (our sim).
• For scale: Earth's normal yearly in-and-out swing is about 3.1 million miles (5 million km).
• The cooling during that second: less than a thousandth of a degree (our calc).

Our take: losing the Sun wouldn't be an instant apocalypse. Nothing happens for 8 minutes, the planets glow for hours on old light, and the oceans buy us months. But the cold never stops, and in the end the only warmth left comes from inside the Earth.

Download our model's output

Plain-text files from the run above. Open the CSVs in any spreadsheet.

• The model's run log (.txt, 11 KB): the script's own printout, with every check.
• Cooling curves (.csv, 28 KB): land, sea surface, ice thickness and the crude global average, warm, middle and cold, from 1 minute to 2 million years.
• Ghost light (.csv, 1 KB): for each planet: when it loses the pull, and how long it keeps shining after our sky goes dark.
• Orbit trace (.csv, 104 KB): Earth and the Moon's straight-line path, plus Mars, Jupiter and Saturn.
• 1-second blink (.csv, 24 KB): how far Earth is off its old path, day by day, for 2 years.

Edited for the web. In the run log we replaced our local folder paths with the file names; the numbers are unchanged.

Images in the video

• The Sun: NASA/GSFC/Solar Dynamics Observatory (public domain)
• The Moon passing in front of Earth: NASA / DSCOVR EPIC team (public domain)
• The Milky Way and Jupiter over Joshua Tree: NPS / Emily Hassell (public domain)
• Saturn: NASA / JPL / Space Science Institute (public domain)
• The Americas at night: NASA Earth Observatory (public domain)
• Hellisheiði geothermal power station: ThinkGeoEnergy / Flickr (source, CC BY 2.0)
• Arctic sea ice: NASA Goddard / Flickr (source, CC BY 2.0)
• Hydrothermal vent chimney: NOAA Vents Program / NOAA Photo Library (source, CC BY 2.0)
• Earth: NASA Earth Observatory (Blue Marble Next Generation); clouds: NASA Goddard
• Jupiter: NASA/JPL/Space Science Institute
• Map data: Natural Earth

Photos are shown unedited (cropped and resized only). The vanishing Sun, the frozen Earth, the switch and everything else are drawn in code by Hourwiz.

Sources

Every fact on this page comes from these pages, opened on Oct 16, 2026. Things change: check the live page before you rely on it.

1. C. S. Baird, West Texas A&M, “What is the speed of gravity?” (2013, updated 2023) It can't really happen; gravity's changes travel at the speed of light.
2. NASA, Sun facts 99.8% of the solar system's mass; about 5 billion more years.
3. JPL Solar System Dynamics, astrodynamic parameters The au, c, light time for 1 au, GM of the Sun, Earth, Moon and Jupiter.
4. JPL, approximate positions of the planets Planet positions on the vanish date.
5. NASA NSSDC, Earth fact sheet Closest and farthest distance, orbital speeds, temperatures, air mass.
6. NASA NSSDC, Moon fact sheet 384,400 km; 1.022 km/s.
7. NASA NSSDC, planetary fact sheet Planet distances from the Sun.
8. LIGO Scientific Collaboration, Virgo Collaboration, Fermi GBM and INTEGRAL, ApJL 848, L13 (Oct 20, 2017) Gravity and light: the same speed; 1.74 s apart.
9. LIGO Caltech, press release on the 2017 neutron-star collision About 130 million light-years; gamma rays about 2 seconds later.
10. LIGO science summary
11. NASA Space Place, why is the sky blue? The sky is scattered sunlight.
12. NASA Space Place, all about the Moon The Moon reflects sunlight.
13. StarDate (University of Texas McDonald Observatory), why aren't stars visible in the daytime?
14. Popular Science, Holly Otterbein (2008, updated Oct 7, 2021) Below 0 °F within a week, −100° in a year; David Stevenson's −400°; the ice blanket; Iceland.
15. D. S. Abbot & E. R. Switzer, ApJL 735, L27 (2011) Earth's heat flow 0.087 W/m²; ice conductivity 651/T; about 3.5 Earth masses. Full text: ar5iv
16. NIST CODATA, Stefan–Boltzmann constant
17. NOAA Ocean Service, can the ocean freeze? About 28.4 °F.
18. FAO, some useful facts about water and ice Latent heat, ice density, sea-ice conductivity.
19. TEOS-10, seawater heat capacity
20. NASA Glenn, specific heats Air. See also the equation of state.
21. Whittington, Hofmeister & Nabelek, Nature 458, 319–321 (2009) How fast rock carries heat.
22. NOAA Ocean Service, sea surface temperature facts The ocean covers 71% of Earth's surface.
23. NOAA Ocean Service, how deep is the ocean? Average 3,682 m (12,080 ft); deepest 10,935 m.
24. WMO, the world's coldest temperature record −89.2 °C (−128.6 °F), Vostok, July 21, 1983. Also Guinness World Records.
25. NIST Chemistry WebBook, nitrogen Boils at 77.34 K. Oxygen: 90.2 K.
26. NOAA, tides tutorial The Sun's tide force is about half the Moon's.
27. NOAA Ocean Today, life on a vent
28. NOAA PMEL, chemosynthesis
29. Deep Carbon Observatory, via Phys.org (Dec 10, 2018) Life deep in the Earth.
30. Samorka, district heating 90% of homes in Iceland heated with geothermal.
31. Thomson Reuters Foundation via the World Economic Forum (Dec 2018) Iceland's winter greenhouses. Also ThinkGeoEnergy.
32. Verkís, Hellisheiði power plant A geothermal heat and power plant.

Our model is a thought experiment: the Sun can't really vanish. Our numbers are rounded as shown, and the cooling model is a simple 1-D model, shown as a range. NASA, JPL, NOAA, LIGO, Popular Science and the other sources didn't make, review or endorse this page or the video.

No sponsor and no affiliate links on this page. Vendors are named so you can check them yourself, not as recommendations.

Follow Hourwiz

• YouTube@hourwiz (opens in a new tab)
• Instagram@hourwiz.ai (opens in a new tab)
• Facebookhourwizhq (opens in a new tab)

Get the week’s videos in one email

All video pages · Back to Hourwiz home