Hacker News Reader

> You can measure and verify that the Sun can't be keeping the Earth this warm, if it was just down to the Sun this place would be uninhabitable.

This is somewhat misleading because the Sun is pretty much the sole energy source for keeping the surface of the Earth warm:

> Despite its geological significance, this heat energy coming from Earth's interior is actually only 0.03% of Earth's total energy budget at the surface, which is dominated by 173,000 TW of incoming solar radiation. [1]

The problem posed by Kelvin was the the interior of the Earth was too warm, which may affect long-term habitability because plate tectonics plays a role in regulating the atmosphere's CO2 levels (or so I've heard), but of course nobody at Kelvin's time knew that. Kelvin wasn't asking "Why isn't the Earth a dead planet?"

[1] https://en.wikipedia.org/wiki/Earth%27s_internal_heat_budget

GP stated

because it's warm underground

They also stated

> if it was just down to the Sun this place would be uninhabitable

which implies surface.

Yup. Bang to rights gov, I was under the impression that insolation isn't enough to do 275K on the surface (ie temperature above freezing) and thus you couldn't really live here, but actually I was wrong.

I think Kelvin meant that if the Sun was keeping the Earth that warm underground then the surface would be frying.

Actually, main thing that kelvin got wrong is that the Earth has a liquid mantle. Convection in the mantle means that the surface receives more heat from the interior and takes longer to cool down to the present temperatures. Radioactive decay inside the earth also adds to the age of the earth but not as much as the convection in the mantle does.

That's fair, a "convective mantel" model which accepts Kelvin's other assumptions gives 2-3 billion years, so it's at least ballpark compatible with biology. I think you'd have struggled to hide the mismatch eventually but it does pass a laugh test unlike 20 million years.

Unlike radioactivity, Kelvin could have considered convection and so that's his fault.

I don't understand how convection can increase the time to cool down. It seems to me that the higher temperatures closer to the surface facilitated by convection would lead to faster cooling.

Faster cooling, but smoother gradients. It means that if you go down a hundred miles, it will be warmer, but the entire center of the earth will be closer to the same temperature. If you assume a solid, you assume a certain constant thermal conductivity, as Kelvin did, even though with a liquid, that's not true. Basically, he assumed that the heat was diffusing slowly, and that it couldn't go above a certain temperature at the core (i.e. not a gas or liquid), and so you had hard limits on the total reservoir of heat in the planet and could calculate a linear fit on it. It turns out the effective thermal conductivity of the center of the earth is much higher than the thermal conductivity of e.g. continental crust.

> If you assume a solid, you assume a certain constant thermal conductivity, as Kelvin did, even though with a liquid, that's not true

I think the liquid is a red herring here. The important thing is a that there are two layers with different properties. Would work just as well with an inner highly conductive solid.

WIthout convection the earth as a whole takes longer to cool down but the surface layer cools faster. In an earth with a convective mantle the earth as a whole cools faster but the surface layer cools slower (because it is receiving more heat from the core).

This article helped me understand that back in the day: www.blc.arizona.edu/courses/schaffer/449/England%20-%20Perrys%20Neglected%20Critique.pdf

The mantle is (mostly) not liquid. It can transmit shear waves, which a liquid cannot do. The convection is occurring in a solid, by plastic deformation.

What's sad is that Kelvin could have used his reasoning to determine that this convection must have been occurring, and from that inferred that continents likely drift around, driven by convection. But he was driven by religious convictions to try to demonstrate evolution couldn't have occurred because there wasn't enough time.

It delights me how scientific discoveries can be so interconnected. It never occurred to me that Marie Curie's work indirectly helped prove Darwin's theory of evolution

Kelvin also publicly stated we would never have heavier than air flying machines.

To be fair to Kelvin, it was a reasonable opinion at the time:

1) Half of the aviation pioneers died trying.

2) The Wright brothers were truly overachievers, devising a control system, early aviation engine (1903) and wind tunnel.

3) Producing light-weight, reliable and powerful aviation engines took over 20 years. The Atlantic crossing only became survivable after the Wright Brothers made the Whirlwind J-5 radial engine (1925):

https://en.wikipedia.org/wiki/Wright_R-790_Whirlwind

For millennia, the pursuit of flying was about as successful as perpetual motion or alchemy.

What's interesting is that it was possible to make gliders for thousands of years. The secret ingredient is to glue fabric along the frame, not sew.

To be fair to his critics: birds, bats, and bees exist.

He didn't say things heavier than air can't fly. Just that we can't make them... (and at the time, we didn't have any success making them - we did however had great success with "lighter than air" air displacement vehicles, namely hot air balloons).

If nature did it 3 times (more, we now know), it’s a bit presumptuous to say humans won’t be able to.

Nature has created conscious beings. It would be a bit presumptuous to assume humans will be able to do so too

Consciousness is clearly a different kind of problem than flight, for one thing it is possible to tell if something is flying buy not possible to tell if it is conscious.

It would be very presumptuous to categorically rule out that humans will create conscious beings. We don’t know how to do it yet but if nature did it, why wouldn’t we?

I disagree, but we've wandered off topic.

Aviation contemporaries made prototypes that either glide or flied powered, sometime poorly.

Assume a spherical planet. OK an ablate spheroid. OK a ...

Now assume Lord Kelvin is talking bollocks. Just because you invent "really cold, bordering on quite nippy" doesn't mean you can overlook things like half lives and red shift. If Lord Kelvin lived now I suspect he might have other conclusions - he evaluated the evidence available at the time.

I was deliberately provocative and I'm glad I was. I've got far better responses as a result. I never thought to mention Kelvin and you have dived in and filled in that gap.

The article in SA was and is absolute bollocks. Bear in mind that your and my definition of bollocks may be dissimilar and I am not being disrespectful. The article is very short and does not provide working. It is of its time and I suspect the working is available somewhere else. It gives us a short glimpse of popular science writing 150 odd years ago.

> I was deliberately provocative and I'm glad I was. I've got far better responses as a result.

I would tend to argue that this isn’t something we need on Hacker News. (Also, it’s “oblate”. To “ablate” is to erode.)

> To "ablate" is to erode.

I mean, it's a ablate sphereoid too, thanks to rainfall, but that's clearly not what they meant.

I think it would be an "ablated spheroid"?