North pole to melt this year?

What is the trend?

Why don't you look at the other data series and longer time frames? You really think that looking at a shorter time frame thereby keeping the confidence level below 95% on the warming trend means something? You can increase the confidence level to 95% by either removing the noise (which was done in the paper that I linked earlier) or extending the time frame. All five sets of temperature data (both raw and adjusted to remove noise) show a relatively consistent increase of around .18 degree Celsius per decade.

You do realize this right? You do realize that the claim there is "no statistically significant warming" is sophistry and does not mean there is no continuing warming trend. The claim that warming "stopped" or is "flat" is just stupid and is really not worth even discussing because I would rather play with my dog.

Here is kind of an interesting article on what you (or really your blogs) are doing:

The Link

And here is another refuting the whole "it hasn't warmed since 1998" nonsense:

The Link

And here is a peer reviewed paper on it:

The Link

And here is a picture:

fawcett_11yr_avg.gif


I am really growing tired of this thread and the whole anti-science pro-stupidity charade. I guess this is part of the plan because the US has pretty much done nothing to curb greenhouse gas emissions for the last 20 years when the benefits would have been greater and the costs less.
 
Paso,

Are you simply obtuse or are you being deliberately deceptive. I am not NOR HAVE I discussed the 1995-2010 time period. I'm talking about the past 15 years which in what you have said has SSW.

You keep attempting to use a different time period to prove your case. That's either a mistake on your part or you're being deceptive or you just don't really understand that proof during one time period doesn't show proof in a different time period.
 
The exact last 15 years with what data set and with or without adjustments to remove noise? And what confidence level?

This is a joke. The warming trend has continued unabated for the last 40 years and it averages right around .18 degree Celsius per decade. I get that with a 15 year or shorter time increment and a certain noisy data set you can sometimes get the warming trend to fall just outside the 95% confidence level (primarily due to the distortion impact of El Nino coupled with an undersampling of the arctic) so you can make the cute claim that there has been "no statistically significant warming". This is sophistry if you know **** about science or statistics. It also absolutely does not mean that there is no warming.

The trend is all that matters and if your trend which is warming (not flat or cooling) is just outside the 95% confidence level, you add another year, try to remove noise, or look at other sets of similar data. This is what a real scientist would do. This is exactly what the peer reviewed paper that I linked earlier did.

I do not know whether you did not read my link or did not understand it, but they absolutely showed statistically significant warming over the last fifteen years. It was made easier by their removal of various distortions (ie noise) and combining of five data sets. This is solid really good science (because this is exactly what good scientists do).

You are not interested in this though. I am, at this point, not even sure what you are interested in. You seemed at one point to be somewhat open minded about this and somewhat knowledgeable about the science and statistics. This no longer appears to be the case. You appear to be interested in playing the mop game of willful ignorance.

You guys can continue to be cute because this has devolved into something way too stupid for me to waste my time with.
 
I think we have our answer AWK. Paso can't work out the trend at all. He appears to be all talk. No one questions that the trend over the past 200 years has been warming. That is obvious. We were in the Little Ice Age 200 years ago, so of course we have warmed. The question is about the past 15 years because it appears the rate of warming appears (contrary to models' predictions) to have slowed considerably. You were putting all these questions to us as if you could do the calculations, so do them. I want YOU to do them. I don't know how. I have said that. Use the data set that Phil Jones and the climate gate crew love the most. That is the HadCrut 3. It is THEIR data set. Show me the statistical warming from the past 15 years.

By the way, if you went back to 1940 and plotted the trend you would not come up with .18 Celsius per decade. It seems it would be closer to .09 per decade so it all depends on where you start doesn't it?

Regardless, we started this segment of the conversation based upon the past 15 years. You INSISTED it had warmed in the past 15 years and implied that we were ignorant for not understanding. So help us understand. Show us HOW much it has warmed since 1997 and what is the trend from the past 15 years.

Or can you do that?

By the way:

1941 through 2011
 
by the way Paso, if you go back 100 years, it actually gets even worse for you…it is about .08 per decade then. But I know you can choose just the right decades and show a warming of .18 per decade. I get that. so what about the past 15 years of data? what are they showing us?
 
so you can't do it can you Paso? I actually thought you had the ability and the know how to work out the analysis of the past 15 years and show us the statistical trend at a 95% confidence level. I guess you are mostly bluster on this.

You were right about one thing though, your response was "wash, rinse, repeat" and really added nothing to the discussion we are having did it?
 
Is approximately .18 C per decade between 1979 and 2010 confusing you? This is at a 95% confidence level. The actual details are in the linked peer-reviewed paper and you can shorten the time frame if you want which they also do in the paper.
Do you need me to teach you statistics? (Don't worry this is a rhetorical question.) You obviously are in way over your head.

This is from the linked peer-reviewed paper:
In reply to:
Click to expand...
Click to expand...
 
paso, you are only embarrassing yourself now. both AWK and myself have repeatedly over and over asked you about the 15 year timeframe from 1997-2011, because THAT is the timeframe about which we made our original claim. i don't question that there has been the trend you have mentioned if we go back to 1979, but that is a different conversation. remember, we don't question that the earth has warmed because it HAS! so trying to convince us that the earth is warmer since 1979 is starting a different conversation. so how about it? can YOU do the analysis on the HadCrut3 data from 1997-2011 as we have requested many many times now and show us the statistical trend upwards with a confidence level of 95%? didn't you claim repeatedly when we mentioned the stand still for the past 15 years that we were wrong and that the rise has been statistically significant? so why can't you show us? you keep picking different timeframes and pretending you have answered our question. it looks like you know no more about statistical analysis than me. why are you cutting and pasting and linking to articles that use different timeframes unless you just can't do the analysis yourself?

it's very obvious what is going on here. I called you out on the carpet and you have been found out. here are the two options from what I have seen.

1. you know exactly how to do the analysis, but see that the last 15 years have had no warming or such a ridiculous amount that you don't want to confirm our point
2. you have no idea how to do the analysis, but were using bravado to bluff your way through the conversation.

i haven't pretended to know how to do statistical analysis, have you?
blush.gif
 
MOP,

You do understand that one reason a regression may be statistically insignificant is a paucity of data points? Fifteen years is a short period for detecting a significant trend in climate data.

Bless you, MOP. You are nothing if not consistent.

texasflag.gif
 
I have linked a peer-reviewed scientific paper that shows statistically significant warming from 2000 to 2010 in every single data set including HadCRUT. I have attempted (in vain no doubt) to explain how short term variations (aka noise) sometimes clutters the signal making it more difficult to show a trend with 95% confidence.

You can adjust for this in one of two ways. The first is to extend the intervals thereby increasing the confidence in the long term trend (ie signal) and the second is to adjust for known items that alter or cloud the signal (which is what they did in the paper that I linked). It does not matter which way you do it; the results remain the same. The earth is warming at a fairly steady pace.

Your blogs try to obfuscate by using the phrase "statistically significant" to mean something that it does not. If a warming trend is not "statistically significant" it merely means that the rise cannot be confirmed above a 95% level to be a non-chance occurrence. Phil Jones honestly answered a very dishonest question. HadCRUT showed an increase over that time frame, but only with a 90% confidence level. What exactly does this mean? I know, but I do not think you do. If you're a serious scientist and are actually interested in what is happening to our environment what do you do? Do you claim this means the trend is flat or do you either extend the time frame or adjust for the noise? I know what I would do (and what the peer-reviewed paper that I linked did).

I have also linked a scientific paper that shows the global warming trend from 2010 back in time to every single year between 1979 and 2005 (including 1997). It all shows pretty much the same thing which is a continuing trend of approximately .018 per year or .18 per decade.

I cannot help your lack of either reading, reading comprehension, or basic understanding. I don't think this stuff is very difficult particularly when compared to climate models (which have always been my primary interest), but maybe this is why you just do not seem to get it.
 
so you can't do it can you Paso? you can't actually do the math and the analysis. you are stuck on "wash, rinse, repeat?" can you not take HadCrut3 data from 1997-2011 and show us that it has warmed? heck, i don't even care if we use the rigorous scientific standard of 95% confidence levels, I am fine with 90%, 85%, 70%! Now I just want you to do the analysis yourself, without the help of highly speculative papers and/or different timeframes. Show us what the analysis is (confidence levels aside) and tell us what the rate of rise is over the past 15 years. do you understand what I am asking? for all of your big talk, it should be QUITE easy to do. you have the data set (it is public domain in its conveniently "adjusted" format) and you have the date parameters. Just do the analysis and tell us what the decadal rise has been over the past 15 years. can you do that? i don't care about the words "statistically significant" any more, just show me the analysis!

how many times do we have to ask the exact same question? are you going to continue to obfuscate? it is painfully, yes painfully obvious what you are doing.
 
yes GT, i actually do understand that. having said that, i don't know if 15 years is too short for that, from what I have been able to gather it is on the threshold of significance in terms of data points. that's why i am fine with lowering the standard, i just want to see if paso can even do the analysis. by this point in the conversation i could have had my best friend do the analysis. he is an engineer who understands statistics well, why can't Paso actually defend his bravado? why can't he put his money where his mouth is?
 
What are you mumbling about? What exactly do you want me to show you? You do realize that this was all done in the paper that I linked right?

Can you specifically tell me exactly what data you want from me? You want like a year by year trend? The trend is approximately .018 C per year. You won't understand the adjustments done to the data. If you want it on raw data, I am going to do it over a longer time frame.
 
MOP, Pasotex has bent over backwards providing links, quotes, & graphs. Now you want him to log onto SAS and do a statistical analysis for you? That's one of the more bizarre requests I've seen on HornFans.

texasflag.gif
 
The saddest part is that I have written down the yearly data since 1995 and am now trying to figure out an easy format to put it in. I think mop is hung up on the 1998 temperature anomaly of .548 because this is the highest one in the entire set (absent an adjustment for El Nino). He does not understand that the next 9 highest temperatures are all between 2000 and 2010. He does not understand what this does to the trend and how you smooth the data set over time. I will try to work on something tonight or maybe tomorrow.

There are actually smoothed graphs on this data showing the long-term trends. I really do not understand why this is not sufficient.
 
paso, this is exhausting. you act as if my request has not been clear when it has been extremely clear. I have even tried to make it easier and easier. for the umpteenth time:

I just want you to give me the HadCrut3 warming trend when the data starting from 1997 (not 1998) to 2011 is included. you have insisted dogmatically that there IS a statistically significant warming trend during THOSE years, so what is it? you can dispense with 95% confidence levels etc. (meaning of course that I no longer care whether or not it is statistically significant so i am letting you off the hook on that!) Just tell me what the TREND is when those 15 years (i.e. "the last 15 years) are included. even if the confidence level is well below the 95% level to be considered statistically significant.

If you can't do that, it's ok, just let us know. Insisting over and over that you have done this already is ridiculous when it is clear to anyone reading (except for GT?) that you have not done it. Telling me that "9 of the 10 warmest years" have been since 2000 isn't helpful either. I already know that. I already grant that we live in a world that has warmed for the past 200 years. I am interested in the past 15 years because it appears the warming has slowed significantly. might it start warming again? sure! but has it warmed for the past 15 years? it doesn't appear to have by very much. so give us the number. clearly it has not warmed by .18 per decade during the past 15 years. that is patently false and easy to observe. So what has the rate of warming been? It is you that insisted that there has been SSW, but now seem completely unable to demonstrate that without switching the years or pointing to speculative papers that try to remove from reality actual events and catalysts. since we don't really know if that paper has succeeded or not, it isn't very helpful to this conversation.
 
Your question tells me that you are not really interested in the last 15 years because if you were you would (1) be interested in the actual trend (which does sometimes need more years or an additional adjustment to remove short term noise like El Nino) and (2) be interested in all the data sets not just one that has a couple of known limitations. You aren't.

I have already posted the data you requested in the link to the peer-reviewed paper (I am pretty sure the lead author is Ramsdorf or something like this but I will relink it if you want me to). You either did not read it or did not understand it or most probably both. I also found the data plotted by a statistician (and yes it shows a warming trend), but I am having trouble figuring out how to display the graphic material. I will try to figure this out later today

I also have the raw data (which is publically available) in the form of yearly temperature anomalies (both actual and smoothed) which I guess I could draw by hand but it would look like ****. Should I just put up the numbers? You obviously do not get statistics or even graphing trends because having one high number in 1998 does not eliminate the trend shown by highs #2-10 being in the 2000's. I knew this without even graphing it because of the math.
 
The internet is awesome. Here is HadCRUT3 data plotted from whatever start time you want to select including 1997. I think the end time is at least 2011 (it sort of looks like it goes into 2012). If you plot it, yes there is a positive slope (ie temperature increase). It is smaller than .18 C per decade. This differs, short term, from the other four primary sets of world weather records.

The Link

If you were actually interested in science and the issue, you would look at why (although actually you would never plot data in such a bizarre way when you have more data than just 1997 to the present). There are two things at work. First, the strong El Nino in 1998 distorts the high temperature quite a bit and more so in HadCRUT3 than the other records. Second, HadCRUT3 undersamples the arctic where the warming is most pronounced. If you extend the years of the sample set and/or adjust for El Nino and the undersampling, HadCRUT3 comports with the other records. There is also HadCRUT4 which incorporates more of the arctic in an effort to compensate for the undersampling (but you are not interested in this).

This is just one more attempt in an endless clown-like effort at sophistry by the denialists.
 
you are hilarious! so we finally find out that you yourself have no idea how to do the analysis eh? which is what i was starting to recognize. no big deal, I don't know how either, but pretending you do is pretty funny and running from all the many requests we made for you to do it is silly.

so, the link you gave me is the one i have REGULARLY posted to, including in this discussion a few days ago. in fact, if i didn't misunderstand you, that's the same site i linked to that you accused of being a "blog." So what gives? do you not check links that people give? I did check yours by the way.

yes…that is an incredibly small trend isn't it? in fact, eyeballing it, it seems to be that over a 15 year period it has gone up about 1/10 of 1/10th of a degree….so .01 over 15 years or .0007 (rounding up) per year. Is it your claim that this is statistically significant? Regardless, I let you off the hook for statistical significance so I will just make another point.

You claimed that the past 15 years has definitely seen warming. You were correct in only the most pathetic use of the term. If "warming" continues unabated as it has for the past 15 years, it would take more than 1,000 years go go up 1 degree centigrade. So yeah, my suspicion that this is not statistically significant is almost certainly true isn't it Paso?

p.s. look back in this very discussion over the past few days to see that I linked to woodfortrees with virtually the same graph you just put up.
 
lest you doubt me, i looked it up. on March 19th at 1:56 in the morning Texas time (i live in Africa), i posted a link to the virtually identical graph to make my point. isn't that when you accused me of linking to a blog or was that a different link i posted that you had in mind?
 
incidentally, i put the graph again on March 20th asking you to tell me how that could be statistically significant. the ONLY difference I can see is that you didn't say 1997 "to 2012" you figured out that the 2nd part was not necessary. therefore our graphs are actually identical!

so let's talk about that .007 decadal trend. frightening isn't it?
 
It is frightening that you think this is a trend or even very useful information. It just reflects your abject ignorance on this subject. You do not get to cherry pick data (both time frame and source) to derive a "trend" particularly when there is lots of additional data both from the same source and from other sources. You also do not get to cherry pick data when there is noise (ie El Nino) distorting the data.

If you do not understand this (which I am beginning to think you actually don't), I cannot help you. I have tried a great deal over the last several pages of this thread, but it does not appear to be working.

I do know how to plot data and I used to do this in undergrad in math, biology, physics, and chemistry classes all the time. I do not know how to do it in Excel or a program that can display it on the internet. I guess I could scan a hand drawing, but this is not very useful particularly when the trend is relatively small and hard to detect.
 
ok, i trust that you do know how to do it, but i am curious as to why you kept obfuscating and changing terms and pointing to papers with different timeframes when we were clearly and repeatedly asking for the past 15 years. and my original claim is almost exactly correct. temperatures have been flatline for the past 15 years. now, i understand that in the larger data set, they represent a higher end of a warming trend. i promise i get that. but if any trend is going to ever reverse it does so by slowing it's rate in one direction and then going flat before actually reversing. i don't think it can yet be determined what is happening currently, but it is safe to say that the past 15 years we have seen precious little warming. .007 per decade is hardly something to be frightened about. in fact, it is smaller than the decades of warming leading up to 1940 when man released CO2 would have been FAR below what we have seen in the past 30 years. so i find it interesting indeed. might it merely turn out to be noise? sure! in 3 years it may be that the warming continues upwards. this would not be surprising as we have been warming for 200 years. until it starts warming at a faster rate (as it appeared to between the mid 70's and the mid to late 90's) consistently, i think the verdict is still very much out. if you look at the jump between 1910-1940 and compare it to the jump between 1970-2000, they are VERY similar indeed.

but no comments about the fact that after over 2 weeks of discussing this you linked to the exact same graph on the exact same interactive site (woodfortrees) as me? you can see why i find this all very amusing. you dodge and duck for weeks and then finally point me back to a graph i pointed out two times on the 19th and the 20th of last month?

As for 15 years, you can call it cherry picking but this was actually the very timeframe that Gavin Schmidt of RealClimate and NASA said would be a concern to him if there wasn't any warming over a 15 year period. so i think it is interesting as well that i am using HIS timeframe of 15 years and seeing that it has been virtually flatlined since 1997. now you can say that the past 10 years have had 9 of the hottest years on record, but that isn't very interesting because we KNOW it has been warming for 2 centuries. so any recent period is going to have higher temperatures than the decades before when we are in a warming trend.
 
Is HadCRUT3 the only source of data for temperatures the last 15 years?

What do the other sources show?

Why would you only use HadCRUT3?

When plotting a "trend", why would you use less data (ie years) than is available?

Are there short term variations that make significantly short term distortions (ie noise) in the long term trend (ie signal)?

Isn't the long term trend all that really matters?

What are these sources of noise and is it reasonable to attempt to eliminate or reduce the noise so the long term trend can be seen?

Does HadCRUT3 contain limitations concerning the arctic?

How would you compensate for this?
 
You must log in or register to reply here.

Weekly Prediction Contest

* Predict HORNS-AGGIES *
Sat, Nov 30 • 6:30 PM on ABC

Recent Threads

Prediction Results Threads

2024 Schedules

Back
Top