Hacker News Reader

Comment by jjoonathan | Hacker News Reader

Are there credible models that suggest Earth will turn into Venus or something? It would take a hell of a lot more to end humanity than it would take to "merely" hurt/kill the most vulnerable segments of our population (which is what I thought the stakes were).

View on HN

No, the Earth has been a lot warmer and colder in the past. We've had massive volcanoes going off for thousands of years, we've had iceball earth, we've had crocodiles and palm trees in the arctic.

The Earth's systems always balance out over the long run. We don't stay trapped in ice ages or hot phases.

The difference is that civilization is a comparatively new development, and nobody really knows how that's going to fare in the face of all this. The human cost is likely to be staggering.

If nobody knows than why do you assume the cost _likely_ staggering?

We're uncertain about the rarity of global collapse due to climate change. The more uncertain we are about our uncertainty of the rarity of a global catastrophe, the more rational it is for us to assume we are likely presently under-estimating its rarity, rather than over-estimating it, because examples of global catastrophe are pretty hard to come by. If nobody knows if it's gonna happen, but the consequence of catastrophe is so bad that it disrupts life as we know it, that means that our uncertainty about our uncertainty of its probability should increase our willingness to defend against its potential ~ -inf payoff, rather than decrease our willingness to.

The thing is, not only is there reasonable evidence that climate change is occurring - which reduces the hesitation to actually spend money protecting against it - but also that it's a problem that concerns pretty much every country.

> The more uncertain we are about our uncertainty of the rarity of a global catastrophe, the more rational it is for us to assume we are likely presently under-estimating its rarity

So the opposite of Occam's razor?

Think of it as Pascal's wager. We are talking about catastrophy here, the cost function is quite out of balance when looking at the alternative of "Reducing our standard of living a bit even though the catastrophy would not have happend anyway".

It took me a long while to come up with a response to this I liked, but think about it this way: there is, maybe, a 10^-a prior climate change destroys life as we know it, right? But my prior for God, heaven, etc existing is, alas, 10^-b, where b >>>>>>>>>>>>> a. (never a 0 or 1 prior kinda guy, even if I know that is dumb in some scenarios, like tautologies and impossibilities, yadada). Too many more factors would have to be revealed to me in terms of the existence of God that I would then believe that a is of similar value to b, if that makes sense? ie that they have similar probabilities (or rather, utilities, which matters since heaven is probably +inf and catastrophe of all life is MAX(badness) which we might as well call -inf, right?

So, not only is -Cost * P(calamity | investment) >> -Cost * P(calamity | not investment), Cost(investment) < [Cost(calamity | no investment) - Cost(calamity | w investment)], so its a rational purchase...

sorry if this is at all unclear, i can clarify anything

Because complexity is the enemy of robustness.

A village that lives entirely off its own subsistence farming, or a band that lives off hunting and gathering, can keep chugging along in merry ignorance of any catastrophe that does not destroy the local food supply (and if one does, they can pick up and move).

A society that depends for daily survival upon massive global flows of food, goods, and information is vulnerable to anything that disrupts those flows.

Even relatively minor changes throw our civilization and economy into panic and disarray. Drought in the Middle East is considered one of the factors that led to the rise of ISIS [0]. But it's not the first time. Drought led to poor harvest which led to the French Revolution when grain prices got too high [1].

If you're wondering why the Netherlands has to reclaim land from the sea, if you wonder where it all went, it was a result of the Little Ice Age and a storm that devastated Northern Europe and washed much the of area out to sea [2].

Of course all of this happened outside of our modern worries about climate change... but it shows even "natural" changes in weather patterns can cause serious economic and cultural upheaval. What we're being warned of right now is far more serious and massive than we've ever seen before, including during all the time periods where drought and famine and storms have famously killed hundreds/thousands before [3].

All you can do is assume the worst. We've seen the impacts throughout history, we're dealing with some of them now. Northern Europe is still paying off the mitigation of the storm that destroyed their land... that storm happened 800 years ago. They'll be paying it off forever.

If we have the ability to stop things like the Little Ice Age or the drought that caused the French Revolution, we should. Right now we're being warned of something like that, but many many times more dramatic. And we have a chance to slow it down. We need to take that chance.

[0]: http://www.politifact.com/truth-o-meter/statements/2015/sep/...

[1]: http://www.academia.edu/4794332/Drought_and_the_French_Revol...

[2]: https://en.wikipedia.org/wiki/Saint_Marcellus%27_flood

[3]: https://xkcd.com/1732/

Yes, absolutely. But the species that inhabit the earth come and go, and climate seems to be one driver.

There's a really good George Carlin bit in one of his stand-ups about how the earth will be fine. Something along the lines of what you're saying; earthquakes, volcanoes, solar flares, etc. She's been around for a looong time folks. It's humanity that might be in for a bad day.

"Save the earth? The nerve, the arrogance. Now the people? The people are f"

There's also this classic page: https://qntm.org/destroy

The difference with the past is that the sun is slowly getting hotter. Increasing the incoming solar power by a mere 10% would be enough to tip the climate into a runaway process that will boil the oceans and scour the surface. Our Earth only has a billion more years to go

https://www.sciencedaily.com/releases/2013/12/131216142310.h...

In a billion years we will find out how to countermeasure this. Is really a long time.

agree, but in the long long run it will lean towards hotter as the sun changes phases. eventually earth will be a scorched rock without anyone's help.

you have to consider, the earth is about 4.5 billion years old, and we only have 1 billion left, but life will be wiped out much 'sooner' then that.

Why does the Earth have one billion years left?

In about a billion years the sun will exhaust enough nuclear fuel to cool and expand, boiling the oceans, blowing off our atmosphere, and turning Earth back into a sterile rock.

It's predicted by astronomers and physicists who study stellar evolution.

No, that doesn't happen until 5 billion years into the future.

The Sun will gradually keep getting hotter and slightly more luminous up to that point.

Earth will be a desert long before the Sun exhausts nuclear fuel to start expanding and becoming a red giant.

All stars become gradually hotter the older they get. G-type stars like our sun are no exception to this. On a certain point in time (currently estimated at 1Gy from now), the energy output of our sun is so big, that the Carbonate–silicate cycle[0] cannot continue uninterrupted. Thus CO2 and CO cannot be absorbed by the geosphere. The breaking of this process will gradually enrich the atmosphere with CO2, leading to a runaway greenhouse effect (a bit like on Venus).

But I would not worry about that. It took 500My for life to evolve from trilobites to humans. This timespan is twice as long...

[0]: https://en.wikipedia.org/wiki/Carbonate%E2%80%93silicate_cyc...

> we only have 1 billion left

Not disputing, but have a source for that theory?

That's less comforting when you realise that cyanobacteria likely caused an iceball phase, nearly wiping itself out in the process.

There is definitely no shortage of doomsday climate models. CREDIBLE models is a different story.

Look back at ones from 20+ years ago to see what they predicted would be happening by now. It's pretty funny actually.

The models from 20 years ago overshot by several confidence intervals because they underestimated mixing the deep ocean. "The barrel is bigger than we expected" doesn't make the rising water in the barrel any less concerning, but it's not the triumph of foresight that you seem to imply.

Grandparent didn't imply it was a triumph of foresight, but rather a failure of foresight.

I think that's what he's saying - a general point that people's understanding of climate (and thus predictions about it) are poorer than commonly believed.

Well, one day we might realize that we underestimated some critical aspect of climate change. Having overestimated it in the past is not an argument for relaxing today, and is certainly not a laughing matter.

Are there any models that predict accurately backwards?

All of them, because that's how you build models? Or do you mean something different?

Pretty sure he ment modern models that prove to be more right than others using past data.

Interesting question imo

I mean per the other response that is how you build models like this aka yes, all of them. Source: Am a Statistician

edit: that doesn't mean they are correct going forward necessarily and most of the doomsday models I've seen are simulations that run thousands of times and a very small number of the simulations predict doom and a very small number predict no effects. The average of these expiriments is often the most likely scenario but doesn't really make headlines.

I don't need models when I have the historical record. The historical record says mean sea level is going to be hundreds of _meters_ above current levels in the not-too-distant future.

This is why people don't listen to the scientists, because absurd FUD like this gets spread and it discredits the entire movement. Even the most doomsday predictions are on the order of inches, maybe over a foot in the next 50 years. Such a rise would still be incredibly devastating, don't get me wrong, but it would take about a hundred years of a completely runaway warming cycle to raise the seas a single meter, much less hundreds.

I'm quite serious about the historical record. This is not FUD, this is data. The historical record shows that the last time atmospheric CO2 concentration was at the level we are at currently, several things were true. 1) temperatures were 4-5C higher. 2) there was no permanent surface ice anywhere on the planet as far as they can tell. 3) mean eustatic sea level was anywhere from 50 to 200m above current levels.

http://onlinelibrary.wiley.com/doi/10.1029/2008PA001615/full

Keep in mind, it's not just melting you need to figure in, there's also the thermal expansion of water.

Also models do not currently project the Antarctic losing all of its ice, yet this is what the historical record suggests is likely. The models also do not account or undercount the contribution of non-anthropogenic forcing factors such as the one this article refers to -- they assume that _we_ are the main feedback loop, and yet there is evidence that we have entered a regime where we could reduce emissions to zero and the non-anthropogenic feedback would continue to add CO2 and methane to the atmosphere at a rate equal to or exceeding our current emissions.

I believe the main reason you aren't hearing things like this is because scientists are picking their most defensible claims because they know that they will be hounded by climate change deniers. They are not talking about their worst fears, for worry of being labeled FUD.

It's not just melting and thermal expansion. There's also loss of mass at the poles, which increases sea level elsewhere.

Edit: sense

You really defended your position quite effectively, thanks for the perspective.

When I read the paragraph 55 of the article, I think that it tells a different story. Basically, it's 20-30 meters higher seas; a good part of it coming from tectonic stuff. And they also say that sea level increase appeared before warming.

The volume of the ocean basin is something that I haven’t been able to find data for one way or the other. The timeline is different — there is a lot of data and of course it doesn’t all line up perfectly. The generally accepted reference timeline, which is based on a single south Atlantic core sample, is calculated using an assumption about the annual rate of sediment deposition. There is evidence that mid ocean sediment deposition dropped during PETM while coastal deposition increased. This is likely because the oceans become inhospitable due to increasing acidity and warming caused the same thing we see today — increased evaporation rate and warmer air being able to hold more moisture as water vapor leads to increased rainfall and therefore river flows increased generally. There have been attempts to try and correct for this factor using iron deposition variability which may correlate with planetary precession and other techniques, some of which change the timeline fairly drastically at the scale we’re talking about.

Long story short — yes this is an oddity but there’s not a well accepted explanation for it and there is reason to question whether we have the PETM timeline correct for the area of interest. (the onset is quite sudden, geologically speaking, although not as sudden as the data we are producing right now!) I would encourage you to do more reading and build an opinion based on the sum of evidence rather than that one paper, which I linked to simply because it has an accessible graph for sea level variability compared to today.

wow that's an answer ! I'm out of my league :-)

> I don't need models when I have the historical record. The historical record says mean sea level is going to be hundreds of _meters_ above current levels in the not-too-distant future.

What in the historical record says this?

Edit: saw your reply to my sibling after I posted this.

http://onlinelibrary.wiley.com/doi/10.1029/2008PA001615/full

Feel free to do your own research, obviously. You are looking for eustatic sea level during PETM (Paleo-Eocene Thermal Maximum).

You will likely find a lot of studies which show coral reefs and anything that lives in the ocean and has a shell is about to have an even worse time of it than we are!

(To be clear, hundreds is the most extreme value -- the variability at different sites is fairly high, but roughly 50m is the minimum.)

I think the worst case scenario is the rising of the sea level by tens of meters over decades, destroying all coastal cities. Think the Syrian refugee crisis, but hundreds of times more people.

> Are there credible models that suggest Earth will turn into Venus or something?

Yes, the model is called Venus.

> (which is what I thought the stakes were).

You were misinformed. The stakes are significantly higher than that.

Venus is a shit model for Earth. Try again.

I’m sorry, but it has long been shown Earth cannot simply turn into Venus given the current climate models.

Reply on news.ycombinator.com