Hacker News new | past | comments | ask | show | jobs | submit

Warm Oceans Turned a 3-Inch Forecast into a Record NYC Blizzard

https://umbrellatoday.app/blog/202602-nyc-blizzard-warm-oceans
loading story #47130093
They don't know what's going to happen tomorrow but they know what's going to happen in 30 years.
A bit of scorn coming your way in the replies but it's not necessarily intuitive if you haven't thought about it. Some analogies that might help:

- If I play roulette in a casino today, I might win big, break even, lose a little or lose a lot. If I play roulette in a casino every day for a decade, I can be nearly certain I will lose a lot.

- Consider an ant walking on a rough stone road built up the side of a hill. If you look at the ant at any particular second, its body might be pointing up (head higher than tail) or down (vice versa) or level, depending on what particular angle of rock it's on at that time. But measured over minutes its likely to be at a greater altitude above sea level than where it started. Measured over the hours it takes to get from the bottom to the top, it's definitely higher.

- A random day of the year (pick from 1-365) in England might be sunny or rainy, but the chances of it being sunnier are higher if the day picked is in the summer.

The point is that there's a tremendous amount of noise in short-term measurements which tend to smooth out over longer term where trends are more clearly revealed. That's the counterpoint to your argument and the reason why climate prediction is not the same as weather forecasting. Going back to the casino analogy, climate prediction is looking at your bank balance over decades; weather forecasting is deciding how to bet on a particular poker hand.

(And finally, we actually kind of do mostly know what's going to happen tomorrow, but not a week out; that's not the point you're making though.)

Are you really equating daily weather predictions with meteorological science? That's like saying "they don't know what the next 3 coin flips are going to be but they know half of the next 10,000 will be tails"
You can’t predict a coin flip because it is random. However, we have an accurate understanding of the random process producing coin flips and therefore, we can make accurate predictions about large quantities of flips.

Weather may or may not be random. It could be entirely deterministic for all we know. However, we lack the ability to fully model all the factors that contribute to weather and therefore our predictions are inaccurate.

Now let’s consider long term climate predications. Do you think these predictions are more like coin flips, where we have an extremely accurate model of the process, or more like weather, where unknown unknowns have outsized impact on accuracy?

That’s not to say climate change isn’t real, but your analogy doesn’t make sense.

> more like weather, where unknown unknowns have outsized impact on accuracy

"Unknown unknowns" aren't the reason weather forecasts are inaccurate.

Weather is path-dependent. Small changes to starting conditions or minor differences between modeled and actual conditions shortly after the simulation begins lead to large differences by the end of the simulation. Errors propagate and magnify.

Over large time periods the errors average out.

All responses are so focused on exact predictions. We have high certainty that 50% of flips will be tails over long enough timespan. We don't know what any single flip will be. Climate science works the same way. But climate is not a coin, let's say it's a multisided die and it appears the sides are changing sizes as we compare data year over year.
loading story #47138406
Unfortunately weather predictions aren’t as simple as a coin flip. But I’m sure meteorologists would manage to fuck up coin flips too.
and humans haven't figured out anything more complex than a coin flip...
That's not exactly true, it seems. Forecasts become less accurate the further out you go, unlike coin flips.

Weather forecasts are generally accurate about 90% of the time for a five-day forecast and around 80% for a seven-day forecast. Forecasts beyond ten days are only correct about half the time.

Why do you pack a light jacket if you go to Tasmania for a week in June, regardless of the forecast?
you're conflating statistics and trends with discrete predictions
Because energy in > energy out is a pretty simple non-chatoic thermodynamic equation with pretty limited variables and weather is one of the most complicated dynamic fractal systems imaginable. Why is this hard to understand? You might as well complain that they haven't described the exact curvature of the coast line of England at the nanometer level and yet they can avoid crashing ships into it with GPS.
And this contrasts with other fields - for example, in orbital mechanics, we're pretty darn certain where the Earth will be tomorrow. We're a bit less sure about 30 years, and increasingly uncertain after that.
If we're talking about dynamic systems then it makes sense
Do you really not understand that this actually makes a lot of sense?
Yes, that's essentially correct. Unintuitive if you don't understand statistics, but correct.