Meaning, cause a seemingly normal winter to suddenly take a turn within hours or a day/two into a completely frozen territory? From what I've read, abrupt climate change happens from ocean currents. Would dramatically shifting the weather happen due to maybe a giant glacier melting up north? I know there's global warming, but I'm looking for a natural "what if" for a story or man-made accident that would speed this up in a way that humans are unprepared for - and how this would affect the world.

Here's what I was reading:


Was reading that the Bering Straight is a 50 mile gap separating Siberia from Alaska, so if that closed, what would happen? It sounds like a more likely idea to use the Meridional overturning circulation (ocean currents) to trigger what's close to an ice age in North America?

Ideas? How can this be explained to me in layman terms? Thanks in advance!

  • 2
    $\begingroup$ I would suggest watching "The Day After Tomorrow" although I wouldn't credit it with too much realism! $\endgroup$
    – Liath
    Commented Oct 7, 2014 at 9:42
  • 4
    $\begingroup$ That film was terrible, like really totally terrible, physics. $\endgroup$
    – Tim B
    Commented Oct 7, 2014 at 13:03
  • $\begingroup$ I feel like I saw that movie - possibly. Maybe it'd be good to re-watch just for the sake of the drama if not the realism of it. $\endgroup$
    – KaguraRap
    Commented Oct 7, 2014 at 16:21
  • 1
    $\begingroup$ I don't want to get all conspiracy on you, but look into chemtrails. $\endgroup$
    – user458
    Commented Oct 7, 2014 at 21:03
  • 1
    $\begingroup$ fictionally, en.wikipedia.org/wiki/Ice-nine $\endgroup$
    – Mazura
    Commented Oct 8, 2014 at 6:49

6 Answers 6


The simple answer is that if you want a permanent change then it is unlikely to happen overnight. There's a lot of inertia in the environment and it takes time for things like currents and suchlike to work.

However there are three candidates that might give what you want, and all act in roughly the same way:

  • A supervolcano
  • A meteor strike
  • A nuclear war

All three of these would act by sending a huge amount of debris, dust and ash into the upper atmosphere.

That dust then blocks and reflects away the sunlight. This kills off all the plantlife and also drops the temperature of the planet rapidly as the sunlight is reflected away instead of warming the planet.

This would normally last at least a few years before it cleared, and it would then take a long time for ecosystems to recover, but the effects in the long run are temporary.

In a hypothetical worse-case scenario a massive meteor strike at one side of the planet can send shockwaves through it. That then causes massive eruptions both at the impact point and at the opposite side of the planet from the impact (as the shockwaves focus again there). If those eruptions were serious and long lasting enough they could extend the resulting cooling for as long as decades or even centuries.

For example the Deccan Traps and the Chicxulub impact:

The release of volcanic gases, particularly sulfur dioxide, during the formation of the traps contributed to contemporary climate change. Data points to an average drop in temperature of 2 °C in this period.

Because of its magnitude, scientists formerly speculated that the gases released during the formation of the Deccan Traps played a role in the Cretaceous–Paleogene extinction event (also known as the K–Pg extinction), which included the extinction of the non-avian dinosaurs. Sudden cooling due to sulfurous volcanic gases released by the formation of the traps and localised gas concentrations may have contributed significantly to mass extinctions. However, the current consensus among the scientific community is that the extinction was triggered by the Chicxulub impact event in Central America (which would have produced a sunlight-blocking dust cloud that killed much of the plant life and reduced global temperature, called an impact winter).

The impact would have caused some of the largest megatsunamis in Earth's history. A cloud of super-heated dust, ash and steam would have spread from the crater as the impactor burrowed underground in less than a second. Excavated material along with pieces of the impactor, ejected out of the atmosphere by the blast, would have been heated to incandescence upon re-entry, broiling the Earth's surface and possibly igniting wildfires; meanwhile, colossal shock waves would have triggered global earthquakes and volcanic eruptions. The emission of dust and particles could have covered the entire surface of the Earth for several years, possibly a decade, creating a harsh environment for living things. The shock production of carbon dioxide caused by the destruction of carbonate rocks would have led to a sudden greenhouse effect. Over a longer period, sunlight would have been blocked from reaching the surface of the earth by the dust particles in the atmosphere, cooling the surface dramatically. Photosynthesis by plants would also have been interrupted, affecting the entire food chain.

  • $\begingroup$ Hmm perhaps a volcano can do it, ever hear of "the year without summer" in 1816? history1800s.about.com/od/crimesanddisasters/a/… $\endgroup$
    – KaguraRap
    Commented Oct 7, 2014 at 6:40
  • $\begingroup$ Yes, that's another example although less drastic than the Deccan Traps eruption. $\endgroup$
    – Tim B
    Commented Oct 7, 2014 at 6:42
  • $\begingroup$ Hmm, I also like how the cause and effect is broken down in your answer: tsunamis, wildfires, earthquakes and so on. I think I may go with super volcano - thanks for the answer. :-) $\endgroup$
    – KaguraRap
    Commented Oct 7, 2014 at 7:34
  • $\begingroup$ @KaguraRap You also have a supervolcano ready to go - read up on Yellowstone en.wikipedia.org/wiki/Yellowstone_Caldera - it's not actually expected to erupt in the next few years and it's not expected to be an extinction level event (for anyone outside north america) if it does but it's going to make a hell of a mess...and you wouldn't need to stretch things much to make it more dramatic. $\endgroup$
    – Tim B
    Commented Oct 7, 2014 at 8:07
  • $\begingroup$ OOH great! I'll look that up although I've been looking into Mt. Tambora so far. Need all the resources I can get. $\endgroup$
    – KaguraRap
    Commented Oct 7, 2014 at 13:46

Something that can trigger a climate change is a sudden change in the orbital pattern around the star. For example, if a gas giant or other massive body passes near the planet, it can disrupt the orbit and either send it into a closer orbit, a further orbit, a more (or less) elliptical orbit or even into the star or out of the system entirely. this isn't as sudden as other events, but depending on the speed and magnitude of change, it can be dramatically different.

This is also something that doesn't have an immediate devastating effect on the population. Nuclear war, a supervolcano or a meteor strike strong enough to have these effecs will also immediately destroy vast areas of your world. A massive body close enough to exert gravitational pull will cause earthquakes and flooding, and potentially some radiation, but it isn't as devastating as the events suggested by Tim B.

  • 1
    $\begingroup$ Nice idea. Something like a rogue gas giant swinging through the solar system would be spotted a good distance out too, there could well be "interesting times" in the run up before it got so close. $\endgroup$
    – Tim B
    Commented Oct 7, 2014 at 9:53
  • $\begingroup$ So not necessarily a meteor but a disruption in orbit can shift climate? I'll look into it. Thanks. $\endgroup$
    – KaguraRap
    Commented Oct 7, 2014 at 13:47

General problem

Note that unless you introduce a massive heat sink, there is an inherent time it takes to cool down the earth or a piece of it because of the following:

We have the climate we have, because the following processes are roughly in equilibrium:

  • Dissipation of heat to space.
  • Absorption of heat from the sun.
  • Absorption of geothermal heat from the earth’s core (this effect is negligible).
  • If you are only looking at a particular area: Absorption from or dissipation to neighbouring areas.

To cool down the whole climate, you have to tilt this balance such that more heat goes out than comes in.

The big issue now is that it’s very difficult to increase the amount of heat that goes out. This may be very surprising at first, because it’s comparably easy to heat things up – you just need to convert some non-thermal energy to heat. However, to cool something down, you cannot just convert heat back to some non-thermal energy, but you have to transfer this heat somewhere (which then heats up)¹. Therefore you either need something very cold or a heat pump (which consumes energy – which is why a fridge in sum heats up a room and you cannot cool a room by leaving the fridge door open). Both are arguably not applicable to your scenario, so you have to rely on the existing large-scale heat drains and the only thing you can do is to attenuate the heat sources and wait.

If the above is to abstract, the following analogy might help: Heat behaves like the water in a lake (with high water level corresponding to high temperatures). There are some rivers feeding this lake (heat from the sun, geothermal heat) and certain rivers draining it (dissipation of heat into space). Right now, the amounts of water going in and out are roughly balanced and thus the water level of our lake does not drastically change. What you want corresponds to draining the lake. However, as you cannot build a new canal to drain the water (introduce a heat sink), you have to wait for the lake to drain via the existing outgoing rivers.

Once more, with numbers

Here are the first two effects in numbers and for illustration:

Thermal equilibrium

(from Wiki Commons)

To take the highest number, we have 343 W/m² incoming heat and 343 W/m² outgoing heat. These numbers being almost equal is the aforementioned equilibrium, which keeps the temperature of earth from drastically changing.

To make a rough and benign calculation, let’s ignore the transfer of heat between areas and assume that you manage to fully switch off the incoming 343 W/m² while keeping the outgoing 343 W/m², i.e., you essentially remove the sun. Further assume, that all you want to do is freeze all the water (i.e., you go from 0°C water to 0°C ice) and that your area is covered by a water column of 1 m on average. We thus have for each square meter:

  • 1 m × 1 m² = 1 m³ of water, which corresponds to 1000 kg.
  • Water’s heat of fusion is 334 J/g. So we need to deduce 1000 kg · 334 J/g ≈ 3·10⁸ J of heat.
  • Going by our assumptions, we deduce 343 W = 343 J/s.
  • Thus, it takes us 3·10⁸ J/343 W ≈ 10⁶ s ≈ 11 d.

So, even under these extreme conditions, it takes eleven days to fully freeze some water. While many assumptions went into this, this should at least give you an idea how long such things take. For comparison, also consider the delay of the seasons with respect to the sunlight: The minimum of solar irradiation actually occurs at what is generally considered the beginning of winter, not the middle.

¹ This is a consequence of the second law of thermodynamics. While there are exceptions to this, they are not applicable on a geological level.

  • $\begingroup$ Hmm that was a bit hard for me to follow but from the gist of it, the climate change wouldn't happen as fast as I'm looking for? And you're saying heat and cooling go hand in hand with one another? Sorry if I'm wrong - science was never my best subject. $\endgroup$
    – KaguraRap
    Commented Oct 7, 2014 at 13:51
  • $\begingroup$ So you're saying that assuming the world is our "lake", if we block out the sun, there's no heat to dissipate and so eventually the "lake" drains? So in weather terms, the planet would cool in general? Haha, am I closer to getting this? XD $\endgroup$
    – KaguraRap
    Commented Oct 7, 2014 at 15:59
  • $\begingroup$ Excellent answer. Another factor would be geothermal heat, we have a lot of hot rock under us that would also slow down global cooling, especially in certain areas. $\endgroup$
    – Tim B
    Commented Oct 8, 2014 at 18:48
  • $\begingroup$ @TimB: That’s what I was referring to with “Absorption of heat from the earth’s core”. However, this effect seems to be orders of magnitude lower, if I did not fully misinterprete the sources. $\endgroup$
    – Wrzlprmft
    Commented Oct 8, 2014 at 19:17

AS Wrzlprmft points out...you aren't going to have a rapid globe freeze or warm-up from an 'earth losses or gains energy' all to readily with a couple exceptions. I won't repeat Tim B's list...except to maybe add a massive space event such as a supernova wave coming our way, or even something off our sun (could have fun there as not only would it impact climate, it'd fry all of our monitoring electronics and make us close to blind)...but I'll give you the redistribution possibility.

The Earth is an incredibly efficient redistributor of energy...in a very short amount of time, energy circulates from the equatorial regions and makes it ways to the polar region to cool. I'll describe 2 main points, though there is a lot more to it:

  1. North Atlantic. Warm water from the Caribbean makes it's way north on the surface of the ocean and makes it's way up towards Europe. The air currents warm on this and blow towards the UK and Europe, providing them with relatively warm trade winds that keep the northern regions relatively warm and rainy, even in winter months.

  2. Pacific. The Pineapple express is a weather pattern (not just a pot reference) where warm air from hawaii and the tropical parts of the Pacific makes its way towards the pacific coast and provides a jet stream of relatively warm and humid air that keeps area's like Seattle and Vancouver very wet, especially in the winter months.

The opposite is true...though I suspect you have some knowledge of that (wunderground is a great site, realclimate is as well) where colder waters descend and make their way back towards tropical regions.

This system is a huge amount of energy in the works...so I'm not entirely sure what it would take to actually halt this system, but in the event it did, I can speculate a few things.

but first:

The storm situation in 'the day after tomorrow' has quite a few flaws in it and the cooling effect wouldn't be so rapid. They explain it as extreme cool air from the upper upper atmosphere being sucked down incredibly quickly by this massive storm...unfortunately that air is also the exact opposite of dense and would warm extremely quickly as it descended. Mountain ranges also have the noted ability to tear apart the air flow in larger systems like this and would prevent the storm systems from forming in their proximity.


air currents move pretty rapidly, however it's not an over night change. It takes around 4 days for particles in forest fires in alaska to blow over canada and end up flowing through canada's tailpipe (the maritimes). Even at it's most extreme, I cannot see this change taking less than a couple weeks, even a couple months.

With that in mind:

Within a week of the system shutting down, water temperatures in northern ocean regions will drop significantly while the regions around the equator would quickly warm up. The UK would probably experience the effect first as the wind changes from Caribbean warmed trade winds to arctic winds coming off a cold ocean. It would happen relatively quickly (within the week time frame) and snow would start to take a permanent grip on the UK landmass. A similar event would start to occur on the west coast of North America as well.

The tropics are a different matter as they would begin to mass heat. There are 2 major factors in hurricane development...ocean surface temperatures and wind sheer. Higher ocean temperatures give hurricanes more energy that result in stronger winds and larger storm surges. I would suggest that hurricanes would be extremely more powerful than we have experienced in the past. That said, I would also expect less of them. Hurricanes are gigantic beasts and stretch from the earths surface to upwards of 60k feet into the air. This creates the scenario where the wind speed in the upper atmosphere can be significantly different than the wind speed at the surface...when this happens, the top of the hurricane that drives the force of the storm is ripped away from the bottom of the storm that has the rotation. So there would be less of an opportunity for hurricanes to form, but when they do watch out.

The end result (2 months?) would be a large icecap over the northern and southern poles and exceedingly warm tropical regions that would hit temperatures outside of what humans consider habitable. The region between the hot and cold would be a massive jetstream of wind that would likely produce sustained windspeeds far greater than what we've seen on Earth so far (I'd go as far as thinking the 'ring clouds' of Jupiter or Saturn could even form).

The only thing that I could consider for a 'abrupt change' is our failing to recognize that this is infact happening (or disagreeing that it is) and we don't realize until the massive hurricane or winter storm strikes. The change would likely be minor at first and not the easiest to detect until the abrupt onset of one of these storms (an arctic storm hitting the UK from the north, or a hurricane of abnormal proportions hitting the Caribbean/US or into the Philippines and china.

Adding in:

Don't neglect an extreme solar event from the list of potentials here. Space weather has a tremendous effect on earth and there is recent research that is linking our weather and storm systems directly to the weather of the sun and the magnetic lashes it sends us. An extreme solar event coming our way could also spawn a series of weather disasters and cripple our electronic ways of monitoring it.

  • $\begingroup$ Note that a rapid warm-up is much easier to achieve (at least with worldbuilder’s methods) than a rapid freeze. $\endgroup$
    – Wrzlprmft
    Commented Oct 8, 2014 at 12:04
  • $\begingroup$ @Twelfth So from what I gather from your answer and Wrzlprmft, is that even if this shift was possible, the fastest it could happen would be a week or two? And the fastest method of the climate shift would be via meteor/asteroid/something obstructing the sun or crashing into our planet? And it's very unlikely that no one would notice - unless I throw in some other factors that would either prevent them from noticing or noticing it too late? $\endgroup$
    – KaguraRap
    Commented Oct 8, 2014 at 16:57
  • $\begingroup$ Yes...I couldn't see it taking less time. Energy moves pretty quickly across the earth, but not too quick. With our current monitoring, we'd probably catch it in 3 or so days as sea surface temperatures are pretty consistently monitored now...I guess you could make the 'denial' case where a group of scientists is making the claim and nobody listens to them until it's been shut down for 2 weeks and has hit a critical point. Prior to 2000, it would have been upwards of a week before anybody really noticed outside a select few. $\endgroup$
    – Twelfth
    Commented Oct 8, 2014 at 17:43
  • $\begingroup$ @KaguraRap - it should be noted that it's not always obvious from measurements what is happening. I imagine a complete shutdown would look like an extreme El-Nino event to some people at first. It should also be noted that if you take the period at the end of this sentence to represent everything we know on the subject, the rest of this post represents everything else there is to know...it's so increbily linked and complex that it's hard to tell what really would happen. Going to edit my post an include one more potential scenario. $\endgroup$
    – Twelfth
    Commented Oct 8, 2014 at 17:50
  • $\begingroup$ @Twelfth Or perhaps if not the denial case, another disaster happening at the same time that distracts from anything else scientists might be bringing up e.g. rapid worldwide plague spread combined with political forces trying to save face and manage the panic from plague - haha would that be overkill? You were talking about air currents in your answer, where would a regional sized asteroid have to hit to ensure deadly cold weather in the easter part of the US, if not just NY? Perhaps that might be a new question I can post... $\endgroup$
    – KaguraRap
    Commented Oct 8, 2014 at 19:36

An extreme low-probability event but possible: two rogue planets collide at the L1 position, the merged mass ends up orbiting at the L1 position. While it won't totally block the sun it would make for a quasi-permanent (it's not exactly stable) partial eclipse.


For a man-made variant of @TimB's super-volcano, the characters of your story might build a large space mirror (for an unobtanium furnace? for a publicity stunt?) at a Lagrange point and accidentally point it at the Earth.
Good ol' xkcd describes an extreme case (which, as so often, destroys the earth). But if you scale it down (maybe a moon- or planet-sized parabolic mirror? I don't have the math to work it out) you would rapidly heat a small area of the earth, with extreme effects.
To be honest, I don't know whether this would lead to a super-cold or super-hot scenario, but it would certainly ruin your day.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .