4 ms·

Just about to post something similar. So much of the global warming rhetoric is just FUD for people who don't understand statistics. For people claiming to know

Just about to post something similar. So much of the global warming rhetoric is just FUD for people who don't understand statistics. For people claiming to know that the increase in temperature for the last 100 years is significant, what is the standard deviation of the change in global temperature over 100 year periods for the last 100k years?

rm999 14y ago
>what is the standard deviation of the change in global temperature over 100 year periods for the last 100k years? I see your point, but we don't have that information and probably never will. We don't have strongly statistically significant measurements until ~1600. But we have to operate with some uncertainty to make decisions worth trillions of dollars. We know the increases we've seen in the past century are of a much higher magnitude than what we saw in the few centuries before that. We believe this is true for at least 1000 years back (see the hockey stick curves). We know the increases in the past 100 years are correlated with human actions, especially greenhouse gas emissions. It is (IMO) common sense to hypothesize this correlation is a causation; we know the causation doesn't go the other way, and we know there are no confounding factors (i.e. the same thing that caused the heating caused us to emit way more greenhouse gases). This is where climate change skeptics attack, which is unfair because it is impossible to prove a causation. If we assume the correlation is a causation, we know the increases we have seen would not have naturally existed. And this is what matters to many scientists. The increases are not "significant" in the statistical sense per se, they are significant in the possible effects on our world.
- zalew 14y ago
  > we don't have that information and probably never will http://joannenova.com.au/2010/02/the-big-picture-65-million-years-of-temperature-swings/ http://joannenova.com.au/2010/02/the-big-picture-65-million-...
  - rm999 14y ago
    It's not useful to compare 65 million years of data + huge error bars with 1000 years + medium error bars with 100 years + tiny error bars. The uncertainty of what was happening 50000 years ago may very well be greater than the changes we are worrying about in the past century (I don't know this for sure, I can't find error estimates in your link).

rm999 14y ago

>what is the standard deviation of the change in global temperature over 100 year periods for the last 100k years? I see your point, but we don't have that information and probably never will. We don't have strongly statistically significant measurements until ~1600. But we have to operate with some uncertainty to make decisions worth trillions of dollars. We know the increases we've seen in the past century are of a much higher magnitude than what we saw in the few centuries before that. We believe this is true for at least 1000 years back (see the hockey stick curves). We know the increases in the past 100 years are correlated with human actions, especially greenhouse gas emissions. It is (IMO) common sense to hypothesize this correlation is a causation; we know the causation doesn't go the other way, and we know there are no confounding factors (i.e. the same thing that caused the heating caused us to emit way more greenhouse gases). This is where climate change skeptics attack, which is unfair because it is impossible to prove a causation. If we assume the correlation is a causation, we know the increases we have seen would not have naturally existed. And this is what matters to many scientists. The increases are not "significant" in the statistical sense per se, they are significant in the possible effects on our world.

zalew 14y ago
> we don't have that information and probably never will http://joannenova.com.au/2010/02/the-big-picture-65-million-years-of-temperature-swings/ http://joannenova.com.au/2010/02/the-big-picture-65-million-...
- rm999 14y ago
  It's not useful to compare 65 million years of data + huge error bars with 1000 years + medium error bars with 100 years + tiny error bars. The uncertainty of what was happening 50000 years ago may very well be greater than the changes we are worrying about in the past century (I don't know this for sure, I can't find error estimates in your link).