Ancient Earth Globe | Hacker News Reader

Ancient Earth Globe

dinosaurpictures.org

405 pointsohjeez72 comments

A couple of links from the last time around:

https://news.ycombinator.com/item?id=24459997

This one has comments/answers by the author that explain his reasoning behind the design (linked to by the thread above - and so we don't need to ask the same questions all over again ;) ):

https://news.ycombinator.com/item?id=17286770

When you search a city it shows its location at the time. I thought it probably just takes in the coordinates but to my amusement, it actually pins the location based on the tectonic plate that the city is on. So it can't show Seattle of 470 million years ago because that tectonic plate didn't go as far back. Intriguing!

How much of that topology is speculation? I get that geologists can estimate the latitude of a location at a given time because of climate and chemical signals, plus some contemporary tectonic shift vectors. I don't get how paleomagnetism is reliable when the magnetic field can wander in the timespan of some thousand years. I'd expect the resulting equations to yield a multitude of solutions or even be unsolvable.

"How much of that topology is speculation?"

Pretty much all of it.

*topography. My degree was in math and topology is a branch of math but when everyone thinks I'm talking about maps :)

Just be cautious asking a cosmetologist about galaxies and star formation, you may be disappointed in the answers.

Not an expert at all but I would expect there to be shared fossil record on coasts that used to be joined up until they are no longer joined. So probably based on that data we can make good estimates of time periods and joined up coastlines.

It would be so incredible if you could hit a Play button and see the plates move (or even just step through the existing snapshots slide show style while centered I your city).

Put it in street view mode, The Time Machine

You can use the left/right arrow keys to step through the existing snapshots, and turn off the "Rotate globe" option in the settings to keep your view in one place.

If you look at the areas with shallow inland seas they line up pretty well with big oil producers today. The shallow sea overlapping with Texas on up to Alberta must have been especially rich in life, considering how our society is still running off energy resources that formed in these seas

Most interesting is tracking Indian sub-continent's journey from Africa to Asia, leaving behind a trail of island clusters at each age.

This is awesome. I like how it goes from 20 million years ago to present in one jump, which is sort of interesting on the timeline. We had a warm period without ice ages 3-5 million years ago, then had the onset of ice ages 2 million years ago(and that cycle might have been terminated by current fossil fuel use). Just a blip in time from this perspective.

Just so we’re clear—the most recent ice age was already ending before the industrial revolution.

Unless you believe like a couple did in the 70s that there was a coming ice age which we completely stopped and reversed. I would love to hear the steelmanning for that.

Not a true ice age, but the so called "little ice age" [1] ends about 100 years after the start of the industrial revolution. So you could argue we helped in its demise, even if it wasn't a "true" global ice age and more of a regional abnormaly that lasted a couple centuries.

1: https://en.wikipedia.org/wiki/Little_Ice_Age

I think all the scientific evidence indicates that we'd be on a long ~50,000 year slide into a new ice age right now, except that human ingenuity in the pursuit of energy resulted in a massive injection of fossilized carbon into the atmosphere in the form of coal, oil and gas combustion.

However, if the goal was to head off the next Ice Age, all the scientists agree, "Mission Accomplished!"

Probably best to eliminate fossil fuels from the energy mix immediately, don't you think?

After 2x-ing the amount of carbon in the whole planet atmosphere in the shortest time ever (ever is a long time btw)? Maybe we could pause and evaluate effects of this geoengineering experiment

We’re still technically in an ice age - there is permanent ice at the poles (for now). In contrast during other ages all the ice melted in summer, even at the poles.

That’s fascinating!

Where can I read more?

Try Wikipedia: https://en.wikipedia.org/wiki/Ice_age

As I understand it, the claim that there was to be a coming ice age was based on our increased rate of particulate matter into the atmosphere, not greenhouse gases. The claim was that in addition to greenhouse gases, we were also exponentially increasing soot, aerosols, and other light dampening particulate matter. They argued that the C02 emissions would increase the temperature, but that cooling particulate matter pollution would outpace the warming greenhouse-gas pollution.

We've very much curbed out aerosol, but increased our greenhouse-gas pollution. I very much doubt that their analysis was correct to begin with, and I doubt had we not curbed the aerosol pollution, we'd be facing an ice age today - but I'm just some jerk on the internet - I could easily be wrong! Right or wrong, the first paragraph above is how I understand the argument to have been made at the time.

[0] https://arstechnica.com/science/2016/06/that-70s-myth-did-cl...

I wonder if there are continents we'll never know about because they get completely subducted and aren't part of any existing continent today.

The map goes back less than a billion years. We don’t know much about the shape of the Earths continents for the first few billion years.

Movement of plate tectonics started probably 750 million years ago. https://www.youtube.com/watch?v=DI6SemRT2iY

Maybe the situation has changed recently, but as I understand there is no scholarly consensus on when plate tectonics started and it is the topic of active debate. For example Onset of Plate Tectonics (2011) https://www.science.org/doi/10.1126/science.1208766 says:

> The further back we look into the geological past, the more obscured the view, masked by an increasingly fragmentary geological record. This has resulted in a controversy on whether plate tectonics operated the same way, or even at all, during early Earth history.

To see life evolve from algae, one would have to live 7.5 million lifespans, assuming one has a lifespan of 100 years. Ancient Egypt is just 50 lifespans away in comparison. That's the most mind-boggling fact I learnt today.

For much of human history 50 years is probably a lot closer. Anyway…

https://xkcd.com/2608

Hah! I was thinking this: https://xkcd.com/505/

As an enhancement to the website it’ll be cool to see how earth might look in future based on the movements so far.

This is a random plate tectonic forecast I found on Youtube: https://www.youtube.com/watch?v=uLahVJNnoZ4

GPlates[1] is related. It has python apis[2], and a web portal[3], with assorted topographic models[4], and another interactive[5] like OP (also based on Scotese's work).

[1] https://www.gplates.org/ [2] https://www.gplates.org/docs/pygplates/pygplates_introductio... https://github.com/GPlates/gplately [3] https://portal.gplates.org/#apps-anchor [4] https://portal.gplates.org/portal/dt/ (select "Mantle Reference Frame" for a similar layout) [5] https://portal.gplates.org/cesium/?view=PaleoDEM https://www.earthbyte.org/paleodem-resource-scotese-and-wrig...

How might we better visually present a heterogeneous block of time?

Consider "today"'s coastline. Earlier today, a mere 15 kya towards Last Glacial Maximum, Florida was twice as thick, and the Boston coast was down past Long Island NY. So how do you non-deceptively show a coastline for "today"? Perhaps use such low resolution that these differences aren't visible? Use an aphysical elevation color scheme which deemphasizes water height? Use timelapse averaging (if a single frame was exposed for 100 ky, then ...)?

Years ago NASA did a global clouds-removed monthly image set. You can see the snow line advancing and retreating with the seasons. See changes in vegetation. Months look very different. What best represents the year? An average of them? The preceding year had different weather. How can the preceding decade be nicely represented? The preceding 100 y, 1 ky, 10 ky?

Clouds are a major visual component our planet. Their patterns change with seasons, with years, with climates, with topographies. How might you show this year's clouds? This decades? This 100 ky? This 1 My?

Climate. Consider the insanely desiccated Pangaean central equatorial desert. In OP, it's colored green based on height. I've seen it shown overlayed with swirls of seemingly cumulus clouds.

Science education graphics have the unfortunate property of combining some aspects done with great care, with many others done with great artistic bogosity, and students left with no way to sort what is which. How might paleoglobe visuals be improved?

I'd love to see a timescale map of the world for the last 10,000 years or so that shows country and empire borders. For example I'd like to be able to see how European country borders shifted during the Napoleonic wars. I haven't found anything like it.

I would love to see the projection of what the landmasses would look millions of years into the future.

I’ve seen some. The most noticeable changes are Australia smashing into Southeast Asia. The Mediterranean closing up as Africa moves into Europe. East Africa splitting along the rift valley. California surfing up toward Alaska.

Will California break off and surf up towards Alaska, you say? Best news I've had all morning.

Something that surprises me about these types of presentations is they all reference the work of Chris Scotese. As far as I can tell that work is over 20 years old from its last coat of paint and 40 years old from when the bulk of the work was done. The website even has the late 90s early 2000s feel (not just in looks but if you click around you can find his kids personal pages from when they were in high school, I wonder what they are doing nowadays). Between the manual artistic license in the renderings to the limitations in simulation and data at the time I wonder how much this is really just like early maps of the world where specific observations are there but largely the picture is drawn more to fill in the gaps than data.

Very true. As the author of this visualization, I airbrushed the country borders off the two most recent textures. Because it was too obvious that the borders show the Soviet Union :)

Professor Scotese was a great partner and instrumental in putting this visualization together. He's acknowledged on the site, but for those interested here is his website: http://www.scotese.com/. I believe he has a more modern iteration of the paleomap that is not downloadable on the web, but for various reasons I did not get those textures in this visualization (they didn't wrap properly iirc).

He also has a nice writeup of the methods used here: https://drive.google.com/file/d/1-q0WIa7ofISFHyBe4UxvN8DIPs8...

I think your "first grass" date is incorrect. Wikipedia says there are grass fossils over 100 million years old: https://en.wikipedia.org/wiki/Poaceae#Evolutionary_history

Pretty much spot on. To summarise, I would say Scotese drew these maps based off ~vibes~, especially when it comes to the paleogeography.

You can see much more cutting edge research here: https://www.science.org/doi/10.1126/science.add2541, which is done within the group that built GPlates. (video of the outputs here: https://youtu.be/QaDt1VQ9WlY )

Very cool.

Like many of these tools it suffers from northern hemisphere bias that hides interesting information about the southern hemisphere.

In looking to correctly simulate axial tilt we get Gondwana and Antarctica hidden from view in the shadow, their sunlit versions only available via options rather than hidden by default.

The southern polar region is the more interesting of the two for the vast majority of geological history and I think as Antarctica thaws we will need more interested geologists in it than our current conversations are generating.

I wish authors would emphasise it more when creating visualisations like this.

Well, if Gondwana wants more prestige, maybe THEY should design a map.

Easy to say but everyone is missing out. Our knowledge is best when shared and we account for our biases.

According to this link, there was no grass during the Jurassic period. But studies show that this is not true. https://www.newscientist.com/article/dn8336-fossil-dung-reve...

THANK YOU!

I regularly think about what the earth looked like X million years ago and I've been wanting exactly this for a very long time :).

This is really interesting, but I have trouble enjoying it because the earth keeps spinning while I try to look at a specific part.

I also have this problem with the actual world IRL

Yeah...astrophotography is a pain because of this.

Go into settings and switch to Flat Earth mode.

Click display options in the top right and then uncheck "Rotate globe"

Are we losing water at a pretty quick pace or something? Seems like there is far more surface area above water now.

Continents can rise and fall without changes in the quantity of liquid water on the surface. One must also consider the fact that the world was much warmer then, so things like Greenland, the Antarctic ice sheet, and glaciers did not exist. If all the ice covering Antarctica , Greenland, and in mountain glaciers around the world of today were to melt, sea level would rise about 70 meters (230 feet). The answer to your question is that a portion of this water turned into ice. Ice that is now turning back to liquid form thanks to climate change. Another factor is that water is also slowly disappearing into the Earth's crust. When the Russians dug 12 kilometers into the crust, they discovered that rocks at that depth were saturated with water: https://en.wikipedia.org/wiki/Kola_Superdeep_Borehole?useski...

How accurate is this? 150 mya it shows the UK and Norway looking almost identical to today.

On a related note, I heard both of those coastlines were formed by glaciers in the last ice age. But wouldn't that also have been true during every ice age? Or did they just get more dramatic each time? ;)

It's showing Tasmania separated from the Australian mainland 20,000,000 years ago, even though supposedly it separated only 12,000 years ago?

Also, the constant spinning is super infuriating and mildly sickening.

You can turn off the rotation in the Display Options.

Separated in the sense land went under water rather than plates moved away?

Very cool website. Would be nice with a slider in addition to the dropdown, so you could scrub backwards and forwards in time.

With the power of a little too much time before a meeting, I give to you a slider;

Add this somewhere which makes sense in the html;

`<input type="range" min="0" step="1" id="mySlider" max="25">`

Then, add this script in your console;

```js

const slider = document.getElementById('mySlider'); const select = document.getElementById('years-ago');

slider.addEventListener('input', () => { select.selectedIndex = Number(slider.value); select.onchange(); });

```

Very cool. Suggestion for the author: provide a key to reset to north-south orientation, as 'R' does in Google Earth.

It's comforting to know there wasn't any ice at the north pole 20Mi years ago, and humans survived just fine.

In hindsight, this is most probably a joke, but I still looked it up.

Earliest humans, Homo Habilis, appeared 2.4-1.4 Mi years ago.

Apes as a whole only split from other monkeys around 24mya, and genus Homo shares ancestors with chimpanzees about 7mya. Evolution happens both incredibly quickly and incredibly slowly depending on what you're comparing it to.

Yeah but the Silurian civilization is gone though :-/

Some time ago I had this thought: The original (single) continent was round, and about the diameter of the moon. Since the moon was much closer to the earth, it's gravitational pull would give rise to a bulge (an earth "tide" if you will), which would be roundish and about the size of the moon's visible diameter.

I am definitely not a geologist, and this is probably un-provable... but, to me, it sure seemed like a plausible idea.

> un-provable ... plausible

Your idea's first hurdle is the length of day vs the length of month. How do you keep the moon over one bulged part of Earth at all times? (it's a very different story for the moon, whose 'day' is as long as its orbit, more at [1]) Additionally, there is a solar tide at work on the oceans that is visible daily, and especially at spring and neap tides. The sun's mass drives a maximum of almost a third of the total ocean tidal bulge.

The second hurdle is viscosity (or if you like, elasticity): tides bulge the most readily-flowing parts of a body the strongest. That means the atmosphere (which is more strongly affected by daily heating <https://en.wikipedia.org/wiki/Atmospheric_tide>), and the oceans are much more strongly affected than the "asthenosphere" (the rocky parts).

In particular, tides bulge the ocean by about a metre. The solid earth tide is closer to 0.2 metres. Thus, the question is: how do you raise a tidal bulge in the rocky part of Earth and keep it from being flooded by ocean tides?

> I am definitely not a geologist

It's more gravitation vs the liquid flow of a stratified (layered) planet than geology that is are the big hurdles.

For experts, tidal Love numbers are the important things for any round stratified body: <https://en.wikipedia.org/wiki/Love_number>, which describe the bumpiness (mass multipole) raised by tidal fores on a spherical body immersed in a tidal gravitational field.

The mantle's rheology (elasticity, viscosity, rigidity), representing about 85% of Earth's total volume, is the primary driver of the Earth's Love numbers. There's more detail on that here: <https://geodesyworld.github.io/SOFTS/solid.htm#link3>.

The crust is quite low-volume by comparison. So a third hurdle is straightforwardly: if there were a persistent bulge in the mantle, why wouldn't a relatively thick part of the crust (a giant continent in your idea) not just "roll downhill", or conversely, how do you get fractions of a round supercontinent to "roll uphill" to some lunar-attraction-induced meeting point?

Finally, with present understanding of continental drift, the continents tend split apart and come together over millions of years, and there's no evidence for anything approaching circularity. The link at the top lets one step through almost a billion years of continental drift, showing this fairly clearly.

- --

[1] Slide 8 ("Tides (2)") of the lecture notes at <https://websites.pmc.ucsc.edu/~fnimmo/eart162_10/Week8.pdf> is pretty accessible, showing an Earth-centric diagram of tides and a moon-centric diagram of its tides, and throughout the rest of the slide deck there's lots of heavy stuff for people who like to grapple with mathematics.

I wondered if it was possible at some point in the past, but I don't think so. The moon's orbit has been increasing its whole existence. It started ~25,000 km away from Earth [0], and is now at ~384,000 km. That means at some point it was at the distance for a current geosynchronous orbit, ~36,000 km. Unfortunately, when the moon was formed, the Earth's day was only 6 hours long [0], so the geosynchronous orbit was much smaller at ~10,000km (based on [1]), and so I don't think they ever lined up.

[0] https://physics.stackexchange.com/questions/31429/how-long-w...

[1] https://www.dummies.com/article/academics-the-arts/science/p...

It's really hard to break global hydrostatic equilibrium of a planet (in fact, that global roundness[1] is used in the IAU's definition of a planet). Any raising of bumps on the surface will tend to produce depressions elsewhere. For example, Mauna Kea depresses the level of the seabed/crust around it. Likewise, high tide at some points (in the solid earth, the oceans, or the atmosphere) are associated with low tides at other points.

- --

[1] You could probably toy with models of Earth-moon as a pair of Jacobi ellipsoids or piriforms (pear-shaped, thin ends inwards) but I don't see that working without a much smaller mass ratio and higher spins. Piriform bodies (at least of homogenous self-gravitating fluid, which is a good representation of the mantle) are generally unstable. Maybe that's good if you can find a path that relaxes back to a Maclaurin (oblate) spheroid for the Earth mass that doesn't also relax the (whole of the) "bump", and relaxes the moon to its weak Jacobi (scalene) spheroid.

Really speculating substantially away from what I know: maybe the "synestia" flavour(s) of the giant impact hypothes(e)s for the origin of the moon might be a path to some test simulation codes: coalesce an ellipsoidal (or as I said, piriform or even oviform) moon first and have that drive some aspects of Earth's planetary differentiation (which happens later in that (family of) model(s): <https://en.wikipedia.org/wiki/Synestia>). In particular, the driving should be away from homogeneity in an attempt to escape eventual hydrostatic equilibrium for the Earth-mass which otherwise leaves you stuck with encoding surface features on the (very) thin crust and then dealing with the Mauna Kea problem above. I don't know how you could approach this idea with realistic chemistry though, which I think melts & dissolves this line of thinking.

ETA: Really wild speculation: with unrealistic chemistry, freeze out a long-term solid hourglass structure with the neck at the Earth's centre of mass, piling lots of rocks on the ends terminating just under the surface (but above the mantle) at the poles, and then have one pole always point to the moon Mass. Doesn't at all fit lots of lines of evidence in very old surface rocks, though. Also very hard to wash out tides raised by the sun.

I want one of those cheesy gemstone gloves but with one these old shapes. Maybe 280 million years ago.

Crazy to think, we had an ancestor that lived on earth during every one of these moments.

(2018)

Keep in mind that:

- the tilt of the earth was substantially different. Where Florida is today, would have been where New York was then. [1]

- the poles (north & south) were in radically different places than today. There’s evidence that dinosuars flourished in what is the artic today because the artic then wasn’t so cold [2]

https://theswaddle.com/earth-had-a-dangerous-axis-tilt-84-mi....

https://www.snexplores.org/article/dinosaur-arctic-bone-foss...

>An April study found that a similar tilt like the one from 84 million years ago is currently happening, courtesy of climate change.

What a completely scitifically illiterate article.

Reply on news.ycombinator.com