Looking at the recent trend won't help us know what the recent trend is like?? Seriously? (Tamino is using a linear trend-line on his graph! How do you think he came up with it??)
It's easy to find out what the temperature trend is since 2000: it's 0.083C per decade, which is so small they can't be sure there is any warming trend at all. Tamino is trying to use a long-term trend to draw conclusions about a short-term trend, which is completely bogus. This is what his graph should look like: http://www.woodfortrees.org/pl...
By the way, your numbers are wrong. (Where did you get.216C/decade from?) The upper bound is closer to 0.165C/decade, which is lower than the warming recorded from '75-'00. But sure, you could argue there is some overlap in the margins of error, and that the trend may have continued. But only if we look to the lower bounds of the margin of error from 1975-2000 and the upper bounds from 2000-2015. If Tamino had actually said that then we wouldn't be having this discussion. Instead he said in unequivocal terms that the warming "shows no sign of slowing down". That is simply not true.
Ironically that's the same argument I made about the "hottest year". Lol. The best you can say is that the trend *may* have continued (however unlikely), and that 2014 *may* have been the hottest year (but probably not).
Maybe you should reread my post. The trend from 2000 - 2015 has not continued as Tamino stated: it is less than half that from 1975 - 2000. If the trend had continued GISS data would show an additional ~0.14C of warming since 2000 and we would not be having this discussion.
What he's doing is comparing the differences between the trend-line from 1970 - 2000 and the trend-line from 1970 - 2015, which are small. That's because 15 years of no significant warming won't effect a long term linear trend-line that much.
Tamino could make the same claim with the RSS data to turn a 15 year cooling trend into warming that "shows no sign of slowing down". It's truly absurd. http://www.woodfortrees.org/pl...
It should be clear by now that Tamino's "clever" approach is misleading and his claim that warming "shows no sign of slowing down" is flat out wrong.
And considering the sparse thermometer coverage, how can they claim to be accurate to within hundredths of a degree?
Oh you are thick. What do you think the uncertainty represents?
I do find it hard to believe that yearly global temperatures can be measured to an accuracy of ±0.05C, as they state on their FAQ. Are they misleading people with that number? Is the uncertainty actually much larger? Do tell.
I've been consistent the entire time. Tamino said: "since the turn of the millenium... global warming has continued and shows no sign of slowing down."
The linear trend from 1975-2000 is 0.018C per year while the trend from 2000-2015 is only 0.0083 per year. That's less than half! So how can he say that global warming "shows no sign of slowing down" since the turn of the millennium?
As I've said before, that approach breaks down rather quickly. What if the highest probability was 10% and the next highest was 8%? That would mean there's a 90% chance it's NOT the hottest year, but by NASA's logic they can still make the opposite assertion.
Of course, this whole discussion assumes the data is good. I don't know what to make of this quite yet, but an adding an extra 2degC to raw temperature data does seem rather high.
He is answering a different question. Again, measuring a trend from 1970 does not tell you if there is a trend since 2000. Nobody denies the long term trend, so what's his point?
I agree he is "clever" by trying to spin it that way. If he had said "there has been no meaningful warming since 2000 but the long term warming trend is indisputable" I could agree with him. But he doesn't say that. Instead he says the opposite and uses tricks to create the illusion of a recent warming trend where there is none.
2010 has a 23% "claim". The chances that another year was hotter is 62%. (That's "more than half".) The AP is now correcting the mistake: http://bigstory.ap.org/article...
Maybe you should explain to them that error bars don't matter.
"Finally"? Nobody is denying the warming trend since the 70's, which is what your graph shows. But there has been no statistically significant warming trend for 17 years or so. That's a fact.
What I said: How can Tamino say "global warming has continued" from 2000 to present when Nasa's own data shows no statistically significant warming during that period?
And now you pretend you agreed all along?
If temperatures went up 10deg one day and stayed perfectly flat for the next 20 years you'd still get a 20 year positive trend. According to your logic you would say therefore "the warming continued" during those 20 years. Good grief.
You are not making much sense anyway. You are saying the upper bound of the margin of error is more than 10 times the trend since 2005??
Assuming you are correct (you're not), I agree we would not be able to rule out a warming trend of 0.254C. Likewise we would not be able to rule out a COOLING trend of 0.23C. Any claims of a positive trend or a negative trend would be even more meaningless.
There's a big difference between what's "important" and what's statistically "meaningful". I say it's not meaningful because it's well within the margin of error, and according to NASA it probably was not the hottest year anyway. If it were outside the margin of error it would be statistically meaningful, but yes, still "not that important". (I'm not sure why you brought it up if you agree it's not important.)
Regardless, I wonder why they didn't bother mentioning such nuances in their press release.
"But collectively we do seem to have an apparent fondness for arbitrary thresholds (like New Years Eve, 10 year anniversaries, commemorative holidays etc.) before we take stock of something. It isn’t a particularly rational thing... but people seem to be naturally more interested in the record holder than in the also-rans."
It's seems clear from the above that the press release was more about PR than it was about science; the impression they left with most people is much different than the reality. Do you think that was by accident?
Of course you will call me a conspiracy nut for not accepting everything the US government says at face value. I think that's naive. You probably really believed there was WMD in Iraq and must now believe it was all just an honest mistake. Otherwise by your own definition you are a conspiracy nut.
Where are you getting your numbers from? (They're wrong in case you're wondering.)
For what it's worth, if you can't rule out a negative trend or a zero trend because of the margin of error, you cannot claim a positive trend. It's not "meaningful". It interests me that you continue to deny something so basic.
The fact remains, the odds are 2014 was NOT the hottest year. Do you dispute that? Apparently you are uncomfortable with uncertainty since you need a clear-cut "winner" when it's clearly a statistical tie. This is not a sporting event.
What if the highest probability was 5% and the next highest was 3%? That would mean there's a 95% chance it's NOT the hottest year, but by your logic they can still make the opposite assertion.
Zeke Hausfather seems to understand what "margins of error" mean. He wrote in 2008 that "the variance in temperature between these [record] years is all within the margin of error of the respective dataset, so year-by-year rankings are not that meaningful." http://www.yaleclimateconnecti...
"Not that meaningful." I guess I shouldn't expect that kind of plain honesty in a NASA press release. NASA scientists seem more interested in politics and staying "on message".
Ae you arguing that since they pick winners in sports regardless of uncertainties, that we should likewise ignore uncertainties in science and simply "pick a winner"? Wow.
So now sports scores are akin to scientific data? But yes, clearly if there were error bars in the Olympics then times that were extremely close and within the margin of error should be considered tied. Otherwise what would be the point of having error bars?
If there is only a 38% chance that 2014 was the hottest year, doesn't it follow that there's a 62% chance it was some other year?
As to RSS, they adjust for such things. But sure, open that can of worms. I'd be happy to start looking closely at all the data adjustments and infilling performed on the GISS temperature data. (Actually I generally don't pay much attention to the two outlying data sets (GISS and RSS). But it goes to show that the data is sacrosanct until it says something you don't like. Suddenly it's not trustworthy.)
IPCC predicted 0.2/decade not 0.02/decade. Get your facts straight.
Once again an AGW proponent tries to re-define "statistical significance" to suit themselves. You wouldn't believe how many times I've seen this happen. It's mind boggling. Anything to win an internet argument, eh? *facepalm*
Q: What temperatures would you have predicted for the last 15 years if warming continued just as fast as the fastest period of warming prior?
Around what the IPCC predicted, about 0.2deg per decade. And no, that's not what we have seen at all.
Gavin Schmidt said there was only a 38% chance 2014 was the hottest year. Take it up with him.
It occurred to me that I should qualify EACH AND EVERY statement with "statistically significant" rather than just the first but I thought there was no way you would be so pedantic since the context should have been obvious.
Regardless, even unqualified the statements are accurate since you can't claim a warming trend if a cooling trend or a trend of zero can't be ruled out. Otherwise skeptics could argue that the earth has cooled since 1997, according to RSS data. (And I'm sure there are some unscrupulous skeptics somewhere making that specious argument.) http://www.woodfortrees.org/pl...
Regarding Hadcrut4 data, there has been no *statistically significant* warming for 17 years, since mid-1997. It will be 18 years of no *statistically significant* warming in about 4 or 5 months. Shouldn't have been hard to figure out if you'd thought about it for half a second, since we're only a few weeks into 2015. But then you would have lost an opportunity to throw mud.
It seems we have reached the point where you are grasping at straws.
It means you can't claim a warming trend because a warming trend of zero can't be ruled out. Grade school stuff, no?
I've seen those graphs, and they show a trend line from 1970 which is not how you determine if a trend exists from 2000. Here are three trend lines. You can see that yes, there is a trend from 1970-2015 that is very similar to the trend from 1970-2000. I've also included a trend line measured from 2000. It is much different.
Linear trends are not very good with curves. If you want to know the temperature trend for the last 15 years you don't start your trend at 1970. As you can see with the RSS data, even 15 years of slight cooling will not effect the long term trend that much. But can you seriously argue that therefore it has continued warming for those 15 years? No, you can't.
You are right, I don't get it. How can he say "global warming has continued" when Nasa's own data shows no statistically significant warming since 2000? His statement is simply not factual.
First, this year MAY have been a record hot year, but likely not. We're talking differences of hundredths of a degree and Nasa forgot to mention the error bars in their big press release...
I said the last 17 years shows no statistically significant warming. To be more precise: the Nasa data shows no statistically significant warming for 15 years, Hadcrut4 shows no warming for 17 years, UAH for 18 years, Hadcrut3 for 20 years, and RSS for 22 years.
No need for bizarre overlapping trend-lines like in that straw-man graphic you posted by Sks. Which skeptics were doing that? I'll tell you: none. They also say skeptics start their trend-lines at the height of the '98 peak to manufacture a negative trend. Again, who does that? No skeptic I know of. Shoddy shoddy shoddy, but AGW proponents swallow it whole-cloth. They just make shit up over there. I'm sorry you fell for it.
Yes, I agree there has been a warming trend since 1970. But it is also true that Nasa's data shows no statistically significant warming since 2000. The data sets differ of course, with Nasa's data showing the most warming and also diverging the most from satellite data. The RSS data for example shows no statistically significant warming since '93 or so.
If ~17 years of no warming is not significant, then it should follow that ~20 years of warming from '76 - '98 is also not significant. (Prior to that human influence was negligible according to the IPCC.) You can't have it both ways. If the warming is meaningful, then so is the pause.
Slashdot Mirror
Looking at the recent trend won't help us know what the recent trend is like?? Seriously? (Tamino is using a linear trend-line on his graph! How do you think he came up with it??)
It's easy to find out what the temperature trend is since 2000: it's 0.083C per decade, which is so small they can't be sure there is any warming trend at all. Tamino is trying to use a long-term trend to draw conclusions about a short-term trend, which is completely bogus. This is what his graph should look like:
http://www.woodfortrees.org/pl...
By the way, your numbers are wrong. (Where did you get .216C/decade from?) The upper bound is closer to 0.165C/decade, which is lower than the warming recorded from '75-'00. But sure, you could argue there is some overlap in the margins of error, and that the trend may have continued. But only if we look to the lower bounds of the margin of error from 1975-2000 and the upper bounds from 2000-2015. If Tamino had actually said that then we wouldn't be having this discussion. Instead he said in unequivocal terms that the warming "shows no sign of slowing down". That is simply not true.
Ironically that's the same argument I made about the "hottest year". Lol. The best you can say is that the trend *may* have continued (however unlikely), and that 2014 *may* have been the hottest year (but probably not).
Maybe you should reread my post. The trend from 2000 - 2015 has not continued as Tamino stated: it is less than half that from 1975 - 2000. If the trend had continued GISS data would show an additional ~0.14C of warming since 2000 and we would not be having this discussion.
What he's doing is comparing the differences between the trend-line from 1970 - 2000 and the trend-line from 1970 - 2015, which are small. That's because 15 years of no significant warming won't effect a long term linear trend-line that much.
Tamino could make the same claim with the RSS data to turn a 15 year cooling trend into warming that "shows no sign of slowing down". It's truly absurd.
http://www.woodfortrees.org/pl...
It should be clear by now that Tamino's "clever" approach is misleading and his claim that warming "shows no sign of slowing down" is flat out wrong.
And considering the sparse thermometer coverage, how can they claim to be accurate to within hundredths of a degree?
Oh you are thick. What do you think the uncertainty represents?
I do find it hard to believe that yearly global temperatures can be measured to an accuracy of ±0.05C, as they state on their FAQ. Are they misleading people with that number? Is the uncertainty actually much larger? Do tell.
http://data.giss.nasa.gov/gist...
I've been consistent the entire time. Tamino said: "since the turn of the millenium... global warming has continued and shows no sign of slowing down."
The linear trend from 1975-2000 is 0.018C per year while the trend from 2000-2015 is only 0.0083 per year. That's less than half! So how can he say that global warming "shows no sign of slowing down" since the turn of the millennium?
As I've said before, that approach breaks down rather quickly. What if the highest probability was 10% and the next highest was 8%? That would mean there's a 90% chance it's NOT the hottest year, but by NASA's logic they can still make the opposite assertion.
Of course, this whole discussion assumes the data is good. I don't know what to make of this quite yet, but an adding an extra 2degC to raw temperature data does seem rather high.
https://notalotofpeopleknowtha...
And considering the sparse thermometer coverage, how can they claim to be accurate to within hundredths of a degree?
He is answering a different question. Again, measuring a trend from 1970 does not tell you if there is a trend since 2000. Nobody denies the long term trend, so what's his point?
I agree he is "clever" by trying to spin it that way. If he had said "there has been no meaningful warming since 2000 but the long term warming trend is indisputable" I could agree with him. But he doesn't say that. Instead he says the opposite and uses tricks to create the illusion of a recent warming trend where there is none.
2010 has a 23% "claim". The chances that another year was hotter is 62%. (That's "more than half".) The AP is now correcting the mistake: http://bigstory.ap.org/article...
Maybe you should explain to them that error bars don't matter.
"Finally"? Nobody is denying the warming trend since the 70's, which is what your graph shows. But there has been no statistically significant warming trend for 17 years or so. That's a fact.
What I said: How can Tamino say "global warming has continued" from 2000 to present when Nasa's own data shows no statistically significant warming during that period?
And now you pretend you agreed all along?
If temperatures went up 10deg one day and stayed perfectly flat for the next 20 years you'd still get a 20 year positive trend. According to your logic you would say therefore "the warming continued" during those 20 years. Good grief.
Again, where are you getting your numbers?
You are not making much sense anyway. You are saying the upper bound of the margin of error is more than 10 times the trend since 2005??
Assuming you are correct (you're not), I agree we would not be able to rule out a warming trend of 0.254C. Likewise we would not be able to rule out a COOLING trend of 0.23C. Any claims of a positive trend or a negative trend would be even more meaningless.
There's a big difference between what's "important" and what's statistically "meaningful". I say it's not meaningful because it's well within the margin of error, and according to NASA it probably was not the hottest year anyway. If it were outside the margin of error it would be statistically meaningful, but yes, still "not that important". (I'm not sure why you brought it up if you agree it's not important.)
Regardless, I wonder why they didn't bother mentioning such nuances in their press release.
"But collectively we do seem to have an apparent fondness for arbitrary thresholds (like New Years Eve, 10 year anniversaries, commemorative holidays etc.) before we take stock of something. It isn’t a particularly rational thing... but people seem to be naturally more interested in the record holder than in the also-rans."
It's seems clear from the above that the press release was more about PR than it was about science; the impression they left with most people is much different than the reality. Do you think that was by accident?
Of course you will call me a conspiracy nut for not accepting everything the US government says at face value. I think that's naive. You probably really believed there was WMD in Iraq and must now believe it was all just an honest mistake. Otherwise by your own definition you are a conspiracy nut.
Where are you getting your numbers from? (They're wrong in case you're wondering.)
For what it's worth, if you can't rule out a negative trend or a zero trend because of the margin of error, you cannot claim a positive trend. It's not "meaningful". It interests me that you continue to deny something so basic.
The fact remains, the odds are 2014 was NOT the hottest year. Do you dispute that? Apparently you are uncomfortable with uncertainty since you need a clear-cut "winner" when it's clearly a statistical tie. This is not a sporting event.
What if the highest probability was 5% and the next highest was 3%? That would mean there's a 95% chance it's NOT the hottest year, but by your logic they can still make the opposite assertion.
Zeke Hausfather seems to understand what "margins of error" mean. He wrote in 2008 that "the variance in temperature between these [record] years is all within the margin of error of the respective dataset, so year-by-year rankings are not that meaningful." http://www.yaleclimateconnecti...
"Not that meaningful." I guess I shouldn't expect that kind of plain honesty in a NASA press release. NASA scientists seem more interested in politics and staying "on message".
Ae you arguing that since they pick winners in sports regardless of uncertainties, that we should likewise ignore uncertainties in science and simply "pick a winner"? Wow.
Clearly they are not making any use of them. Oh wait, they are.
"Sixteen years into the mysterious ‘global-warming hiatus’, scientists are piecing together an explanation." http://www.nature.com/news/cli...
Maybe you should explain to the journal Nature why they should have ignored the margin of error and claimed a warming trend.
The actual trend is as high as 0.254C/decade for GISS
No, that is not the linear trend GISS data shows for the last 15 years. Where did you get that number?
Yes, according to your logic, error bars are not needed. I wonder why they have them. :P
So now sports scores are akin to scientific data? But yes, clearly if there were error bars in the Olympics then times that were extremely close and within the margin of error should be considered tied. Otherwise what would be the point of having error bars?
If there is only a 38% chance that 2014 was the hottest year, doesn't it follow that there's a 62% chance it was some other year?
As to RSS, they adjust for such things. But sure, open that can of worms. I'd be happy to start looking closely at all the data adjustments and infilling performed on the GISS temperature data. (Actually I generally don't pay much attention to the two outlying data sets (GISS and RSS). But it goes to show that the data is sacrosanct until it says something you don't like. Suddenly it's not trustworthy.)
IPCC predicted 0.2/decade not 0.02/decade. Get your facts straight.
Once again an AGW proponent tries to re-define "statistical significance" to suit themselves. You wouldn't believe how many times I've seen this happen. It's mind boggling. Anything to win an internet argument, eh? *facepalm*
Q: What temperatures would you have predicted for the last 15 years if warming continued just as fast as the fastest period of warming prior?
Around what the IPCC predicted, about 0.2deg per decade. And no, that's not what we have seen at all.
Gavin Schmidt said there was only a 38% chance 2014 was the hottest year. Take it up with him.
It occurred to me that I should qualify EACH AND EVERY statement with "statistically significant" rather than just the first but I thought there was no way you would be so pedantic since the context should have been obvious.
Regardless, even unqualified the statements are accurate since you can't claim a warming trend if a cooling trend or a trend of zero can't be ruled out. Otherwise skeptics could argue that the earth has cooled since 1997, according to RSS data. (And I'm sure there are some unscrupulous skeptics somewhere making that specious argument.) http://www.woodfortrees.org/pl...
Regarding Hadcrut4 data, there has been no *statistically significant* warming for 17 years, since mid-1997. It will be 18 years of no *statistically significant* warming in about 4 or 5 months. Shouldn't have been hard to figure out if you'd thought about it for half a second, since we're only a few weeks into 2015. But then you would have lost an opportunity to throw mud.
It seems we have reached the point where you are grasping at straws.
It means you can't claim a warming trend because a warming trend of zero can't be ruled out. Grade school stuff, no?
I've seen those graphs, and they show a trend line from 1970 which is not how you determine if a trend exists from 2000. Here are three trend lines. You can see that yes, there is a trend from 1970-2015 that is very similar to the trend from 1970-2000. I've also included a trend line measured from 2000. It is much different.
http://www.woodfortrees.org/pl...
The same thing happens with the RSS data but more exaggerated: http://www.woodfortrees.org/pl...
Linear trends are not very good with curves. If you want to know the temperature trend for the last 15 years you don't start your trend at 1970. As you can see with the RSS data, even 15 years of slight cooling will not effect the long term trend that much. But can you seriously argue that therefore it has continued warming for those 15 years? No, you can't.
You are right, I don't get it. How can he say "global warming has continued" when Nasa's own data shows no statistically significant warming since 2000? His statement is simply not factual.
First, this year MAY have been a record hot year, but likely not. We're talking differences of hundredths of a degree and Nasa forgot to mention the error bars in their big press release...
I said the last 17 years shows no statistically significant warming. To be more precise: the Nasa data shows no statistically significant warming for 15 years, Hadcrut4 shows no warming for 17 years, UAH for 18 years, Hadcrut3 for 20 years, and RSS for 22 years.
No need for bizarre overlapping trend-lines like in that straw-man graphic you posted by Sks. Which skeptics were doing that? I'll tell you: none. They also say skeptics start their trend-lines at the height of the '98 peak to manufacture a negative trend. Again, who does that? No skeptic I know of. Shoddy shoddy shoddy, but AGW proponents swallow it whole-cloth. They just make shit up over there. I'm sorry you fell for it.
Yes, I agree there has been a warming trend since 1970. But it is also true that Nasa's data shows no statistically significant warming since 2000. The data sets differ of course, with Nasa's data showing the most warming and also diverging the most from satellite data. The RSS data for example shows no statistically significant warming since '93 or so.
http://www.woodfortrees.org/pl...
If ~17 years of no warming is not significant, then it should follow that ~20 years of warming from '76 - '98 is also not significant. (Prior to that human influence was negligible according to the IPCC.) You can't have it both ways. If the warming is meaningful, then so is the pause.
Pardon me, it was not you who was calling people names.