The freezing point of water is the basis of the Celsius temperature scale, marking zero degrees there, while the boiling point of water marks 100 degrees Celsius. In absolute temperature, water freezes at 273.15 Kelvin and boils at 100 Kelvins above that. But what if, I thought today, we made the freezing point of water itself a unit? Let's call it an aqua and work out some notable temperatures in these units, which is done by simply dividing their absolute temperature in Kelvin by 273.15. And, as a callback to this post, I'll show the temperatures as images where each pixel represents one Kelvin and then one aqua (273.15 K).

Absolute zero, 0 K and 0 aquas, is the lowest possible temperature. Since temperature is a measure of the average kinetic energy of particles in a material (usually atoms or molecules), at 0 K they practically stop moving. This temperature is actually impossible to achieve, since it gets harder and harder to take kinetic energy out of an object as you get closer to absolute zero. Even outer space isn't at 0 K despite popular belief - the few harmless particles that are present in the vacuum of space actually move quite fast, technically making space hotter than lava or even the surface of the Sun.

1 Kelvin equals about 0.0037 aquas or -272.15 °C. An example of a 1 K environment is the outskirts of the Boomerang Nebula, the coldest known place in the Universe. The carbon monoxide in that nebula is three times colder than even the cosmic microwave background, which carries energies corresponding to a temperature of 2.7 K (-270.4 °C).

Hydrogen freezes below around 14 K or -259 °C, or 0.05 aquas. The average temperature on Pluto is 44 K, -229 °C or 0.16 aquas. The lowest air temperature ever recorded on Earth, in Antarctica, was 184 K, -89 °C or just over 2/3 of an aqua. Here's how all these temperatures look as pixels, including the aqua itself:

Room temperature (20 °C) is about 1.07 aquas. Water boils at 100 °C or 1.37 aquas. Just two aquas (273.15 °C or 546.3 K) is already hotter than a running oven. The temperature on Venus, which is enough to melt many metals, is about 2.7 aquas (740 K). Lava is usually around 5.5 aquas (1500 K). The coldest known M-type stars have surface temperatures of just over 7.3 aquas (2000 K), which is almost hot enough to boil lead. The surface temperature of our own sun is about 21.2 aquas (5780 K). Even though this temperature is enough to boil many metals, it doesn't seem all that hot now - and this is just absolute temperature in slightly unconventional units. Here's most of that in pixels, with aquas represented by blue squares of arbitrary size:

The surfaces of stars in the Eta Carinae system reach about 150 aquas (I'm rounding things up here since the temperatures of distant stars can get weird). A hotter-than usual day in McKinney is 207 aquas. 100,000 K corresponds to 366.3 aquas and is found on the surfaces of some of the hottest white dwarfs. This was also the temperature of everything approximately 500 years into the Big Bang.

Beyond this point we have to reduce the scale so that one pixel represents one aqua.

One million Kelvin is 3663 aquas. Ten million is 36,630 aquas. The temperature at the core of the Sun, which is high enough to fuse hydrogen nuclei together into helium, is on the order of 60,000 aquas. 400,000 aquas or just over 100 million K is enough to fuse helium into carbon and oxygen; the collapsing cores of red giant stars (such as our sun a few billion years from now) reach this temperature.

We have to drop the pixel visualisations here - we can't zoom out too much without completely losing our sense of scale (does that phrase make sense if we're talking about temperature?..). One megaaqua - one million aquas - equals, of course, 273.15 million Kelvin; there aren't really any examples of this temperature beyond dying stars, fusion reactors and the Universe when it was mere minutes old. A square representing one megaaqua would be 1000x1000 pixels.

When the Universe was one second old, its temperature was about 40 megaaquas - that square would be another 6 times larger in side length. At this temperature, the energies of matter become so great that electrons and positrons are spontaneously produced from light, and neutrinos begin actively interacting with other particles. Beyond the Big Bang, temperatures like this are thought to be found only in the most energetic events, such as gamma-ray bursts.

There are temperatures at which even more bizarre things would happen, and you'll often hear that the Universe had those temperatures in the first second of its existence, but it's kinda complicated - there's no direct evidence of what the time before 40 megaaquas was like (frankly, we're not even sure how long it lasted). In practice, temperatures in the gigaaquas (billions of aquas) are only known to be found in particle colliders or high-energy cosmic rays.

Protons and neutrons decompose into a mush of quarks above 4 gigaaquas. The highest temperature ever produced by humans was 20 gigaaquas. Beyond this point we can only talk about theoretical conditions. At four trillion aquas the notion of mass disappears as the particle responsible for it, the Higgs boson, stops working the way it normally does. Light also stops being a thing as electromagnetism and the weak nuclear force merge into a single electroweak force.

At ten million times that temperature, or forty quintillion aquas, particles of matter as we know it may stop existing entirely, dissolving into whatever quantum froth they emerged out of in the first moment of creation. But, of course, that doesn't stop physicists from going even further.

Our last stop is the Planck temperature, beyond which current physics cannot say what could possibly happen. This is about 142 billion trillion trillion Kelvin or 520 thousand trillion trillion aquas. A cube of that many pixels (well, voxels) would be over 2 million kilometres on the side. Utterly incomprehensible, but a good place to stop messing with the magic thermostat.