Earth impact probability has fallen to 1.5%
chance this asteroid could impact the Moon. Current calculations estimate this impact probability to be 0.8%.
0.28% Earth impact
1% Moon impact
The power of observation
Leave it to nerds obsessed with telescopes to name themselves the PLANETARY DEFENCE TEAM
They didn’t go for a cool acronym? Like S.P.I.D.E.R. (Special Planetary Identification Defense & Eradication Regiment)?
Lame
It jumped slightly to 0.36% two days in a row, but now it’s 0.005%
At some point it will be either 0% or 100%.
Earth has a rounded shape, through hydrostatic equilibrium, with an average diameter of 12,742 kilometres
Wait what’s the meaning of having the probability when it changes every few seconds?
What changes every few seconds? They make observations every night, and once a day they update their estimates based on the new data of the likelihood the asteroid will collide with earth in 2032, or any of its other orbital periods.
I mean, if yesterday it was 3%, and today it becomes 2%, then that means yesterday’s 3% was wrong, and today’s 2% may also be wrong since it may become another number tomorrow. So there doesn’t seem to be any meaning in the estimate anymore until the day eventually comes since it just keeps changing.
Unless I see it like poker. Initially, I have a certain percentage to win based on my hand. But once I see the flop the percentage changes. It changes further when the river comes and so on. But no matter what the percentage is all that matters is when the players show show their hands.
Think of it like a new go rating. When someone first joins the server, or in this case a new asteroid is discovered, we don’t know their rating. As the player completes games, and as we make more observations of the very dark relatively small asteroid flying away from us, our uncertainly goes down and our predictions become more accurate.
That doesn’t mean our early predictions have no value at all, it just means we revise our opinion as new information becomes available.
That’s an interesting way to think about it but…
How do we know that the predictions become more accurate? Couldn’t it be because of data anomalies such as the player being on a losing streak due to being tilted or drunk?
Because a) there are more obervations every day and b) the observations get easier and hence more exact every day, because the object is geting nearer.
All analogies break down at some point.
When modelling the behaviour of two individuals playing on a go board, there is a lot of hidden information and external influences that the model doesn’t account for.
But with an asteroid moving in the gravitational field of the solar system, there basically are no other influences on its course. So when you know its position and velocity, you have sufficient information to predict its course. (That’s why, from ancient times til today, astronomy has been of great importance for the development of mathematics and physics).
With more measurements the accuracy of the predicted course increases, so the accuracy of the probability that it will hit the Earth increases.
This is the fundamental critique made e.g. in the book
https://g.co/kgs/oEVCbV2
of things like “superforecasting”
https://g.co/kgs/Y1si6KT
In principle the kind of prediction made about an asteroid should be possible, it’s just maths and the underlying nature of the system is not changing unexpectedly. However, applying a probability makes no sense because even after the event we won’t know if the prediction was right or not. And we are not going to have many of the same asteroids to make probability meaningful.
In the case of the asteroid it’ll either hit or not as jlt says. People are keen to understand these uncertainties in terms of probability. But this is misleading. What people really want is reassurance that some catastrophe won’t happen (because the probability is small) when really we (humanity) should be thinking about each alternative outcome and preparing for them both. That’s the case for asteroids or more earthly uncertainties like things building up in 2025!