4 ms·
Over 10 years ago, James Hansen called future extreme events "Rolling a Thirteen". You were suspecting the two dice you were rolling were not regular anymore, b
by Gravityloss 5y ago
Over 10 years ago, James Hansen called future extreme events "Rolling a Thirteen". You were suspecting the two dice you were rolling were not regular anymore, because the average had crept upwards, but after rolling a thirteen, you know.
EDIT: It seems to actually have been Steven Sherwood in 2010. https://greenfyre.wordpress.com/2010/08/21/tumblin-dice/ https://greenfyre.wordpress.com/2010/08/21/tumblin-dice/
- teekert 5y agoThere is that and then there is the fact that measuring for increasingly longer time periods will give you more and more extremes. With two dice you usually start by rolling some 7's, a 10, etc. Then after many tries you will have rolled a 2 and a 12. Edit: Note that this is not me saying I don't believe in climate change people!
- syops 5y agoThis is only true if the set of possible outcomes has remained static and if the probabilities of each outcome has remained static. With climate change the probabilities have shifted and the possible outcomes may have shifted too.
- xboxnolifes 5y agoThey may have, but the question is whether or not the assumption that we have been rolling 2 6-sided die is correct in the first place. For all we know we could have been rolling 3.
- syops 5y agoI truly don't understand what your point is. If, in your parlance, we have been rolling 3 dice all along then this means our previous 1000 years of climate data is truly extraordinary since this is the first time seeing a number greater than 12. So extraordinary that it is not plausible to think that we have been rolling 3 dice all along. Of course this is possible in a similar way that it is possible that an unseen intelligence has been controlling things all along. The point about seeing 13 is that the set of outcomes have changed. We are beyond the realm of thinking that nothing has changed. It's not the we have always been rolling 3 dice; it's that we are now rolling 3 dice.
- blackbrokkoli 5y agoYes, but we have an awful lot of new high points in recent times. The earth is quite a big place and we are recording temperatures quite detailed since the 1850ies, so if we are accepting the null hypotheses of there not being a rise in averages, stochastic is rearing it's ugly head quite a bit since the early two-thousands...
- d0mine 5y agoIt is not like that: it is not about how long you look at it, it is about "one in a century" events are happening once in ten years, or even once a year now.
- teekert 5y agoYes, but, we have been measuring (and at increasing accuracy and with a changing landscape) for not so very long.
- imoverclocked 5y agoTrees have been measuring far longer. Other indicators of climate go back a very long way and we have a pretty good idea of major shifts in the past. It’s not like we just create data from nowhere…
- xadhominemx 5y agoThe longer time goes on, the more rare records should become. But record highs are becoming more common
- jansan 5y agoIn Canada most temperature extrema are still from the first half of the 20th century: https://en.wikipedia.org/wiki/List_of_extreme_temperatures_in_Canada https://en.wikipedia.org/wiki/List_of_extreme_temperatures_i...
- xadhominemx 5y agoAny very large deviation dating from the 20s or 30s is probably a measurement error
- phreeza 5y agoI think you have it backwards. For a stationary process, the probability of getting a new "all-time high" actually goes down as your record-keeping time increases.