Water is necessary for supporting "life", at least the particular form of life as we know. The origin of Earth's water is still debated, but there are clues that most of it was delivered, later after the creation of Earth Wiki. It is now established that other planets in our solar system harbor water, in at least one of its three states (solid, liquid, vapor). So, Earth may not be the first planet in the solar system to have liquid water. Comparing the age of the Sun to the age of the Universe, there is strong suspicion that liquid water may existed on exoplanets long before the Sun was formed.


I'm curious as to: in which time frame the very first planets with liquid water were likely to have existed, and how this is scientifically established/hypothesized/modelized?

Subsidiary: What is the odd for an exo-planet to harbour water in all 3 forms, like on Earth?

  • 3
    $\begingroup$ can you add some more context to this question, such as "I have search on the internet and discovered.... so I'm curious ..." $\endgroup$
    – James K
    Jul 21, 2021 at 8:33
  • 5
    $\begingroup$ Do you mean in the universe? In the Solar System? $\endgroup$
    – ProfRob
    Jul 21, 2021 at 9:06
  • 2
    $\begingroup$ The interesting thing is not "WHEN", but HOW did they (the scientists) establish the time liquid water first appear, on Earth, Mars, etc. (anywhere else that they have proved its existence now, or in the past). If we don't understand (or believe in) the HOW, the WHEN becomes moot. $\endgroup$
    – Ng Ph
    Jul 21, 2021 at 14:03
  • 1
    $\begingroup$ @NgPh I've added a reopen vote to get the process started. $\endgroup$
    – uhoh
    Jul 23, 2021 at 23:27
  • $\begingroup$ I've voted against reopening, because two key details are still missing: where (just in the local stellar neighbourhood? Just in the Milky Way? In very distant galaxies?) and when (e.g. in recent times, i.e. of exoplanets we can actually observe, or at any time in the history of the universe, i.e. including planets whose host stars have long since ceased to burn)? Also, the recent edit has added an entirely new question, about all 3 water phases being present: this should be deleted, and asked separately. More editing please... $\endgroup$ Jul 24, 2021 at 4:31

2 Answers 2


The question of water delivery to Earth, and planets in general, is one that is hotly debated!

Let's start by using Earth as our benchmark. The total amount of water on the Earth's surface is around ~1.5 x 1021 kg, and we refer to this amount as one 'ocean' of water. There's also a huge amount of water inside the Earth itself, locked in e.g. hydrated silicates and in the core. We don't know how much this is, but estimates range from about 1-50 'oceans' worth. So any models that attempt to describe water delivery already have a problem because we don't fully understand how much water there actually is!

One important feature of Earth's water is that it has a distinct D/H ratio (the ratio of deuterium to hydrogen), which is ~6 times higher than that of the Sun and the gas planets in the Solar System. We can use this fact to constrain where exactly in the Solar System Earth's water might have come from. Current theories suggest that carbonaceous chondrites are a likely source.

So let's take a look at some of possible delivery methods during Earth's formation within the protoplanetary disk:

Direct accretion of water from the protoplanetary disk

Water may have been accreted directly from the disk in the form of water vapour adsorbed on to silicate grains. This scenario could have seeded the Earth with a few 'oceans' of water, but doesn't account for the high D/H ratio.

Alternatively, a planet may accrete water in the form of solid pebble 'snow' if the planet exists sufficiently far away from the star beyond the water snowline. This is not the case for Earth, but is certainly an important mechanism for planets in general. It is important to consider the fact that protoplanetary disks are dynamic systems in which the snowline isn't static. Depending on the conditions, a static planet could see the snowline sweep past it during the course of its formation.

Reactions with Earth's early magma ocean

Gravitational accretion of hydrogen from the protoplanetary disk may have reacted with the early Earth's magma surface to generate water. This method also predicts a low D/H ratio, but can be accounted for somewhat by a process called 'fractionation', which increases the D/H ratio with time.

Scattering of material by other planetary bodies

The growth and migration of giant planets can cause chaos within a protoplanetary system, scattering and perturbing the orbits of smaller planetesimals. In the case of the Solar System, the growth and migration of Jupiter is thought to be responsible for the inward scattering of planetesimals, trapping them in the asteroid belt. Others were scattered further in, and could potentially have delivered water to the terrestrial planets.

Now, to answer your question... When did the first liquid water settle on planets? Well, the answer is that we don't know, because we still don't know how water was delivered. Finding the answer to this is one of the major ongoing areas of research in planetary science.

For further reading I suggest these two excellent recent reviews:
Oberg & Bergin (2020) - "Astrochemistry and Composition of Planetary Systems"
Raymond & Morbidelli (2020) - "Planet Formation: Key Mechanisms and Global Models"

  • 1
    $\begingroup$ I like this answer: let's establish clearly first what we don't know. $\endgroup$
    – Ng Ph
    Jul 22, 2021 at 8:16

If the question is when did the first planets in the universe have liquid water, the answer is billions of years before the Solar System formed.

The Earth is about 4,600,000,000 years old, and I think that the geological record shows that liquad water was already present about 4,000,000,000 years ago. So Earth gained liquid water less than 1,000,000,000 years after it formed.

The age of the universe is estimated to be about 13,000,000,000 years. This indicates that there could possibly have been planets with liquid water about 1,000,000,000 years after the universe formed or about 12,000,000,000 ago.

But it would take longer than that, because the heavier elements that planets are made of would have to be formed by nuclear fusion inside large stars which would then become supernovae and spread their heavier elements thorugh space. Eventually the gas and dust in interstellar space would have a high enough percentage of heavier elements that planets could form when stars formed.

So the first planets with water might have appeared billions of years after 12,000,000,000 years ago. Perhaps the first planets with liquid water formed about 10,000,000,000 years ago, more than 5,000,000,000 years before the Solar System and the Earth formed.

  • 1
    $\begingroup$ This is a good reasoning to give a rough time-scale perspective, but the answer should be "there are good reasons to SUSPECT that it happened long before the Solar System is formed". $\endgroup$
    – Ng Ph
    Jul 22, 2021 at 8:09
  • $\begingroup$ There seems to be a lot of guesswork here, and no authoritative evidence is cited. You seem to assume that "heavier elements" are required to form a planet – which is obviously true for rocky planets but you offer no reference to support the case for early-universe gas giants. You also seem to discount early stars being more massive & hotter, burning up fast (say, 10 Myr) & seeding the galaxies with metals at a very rapid rate. PS using numbers with so many zeros makes it very difficult for the reader: better to use "billion years" or "Gyr" or "$10^9$ years" (etc). :-) $\endgroup$ Jul 24, 2021 at 4:19

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