Guest Post by Werner Brozek Edited by Just The Facts:
To make this discussion easy, I will make the following assumptions. Dr. McKitrick’s data went until April, 2014, however I will assume his data continued to October, 2014. I will assume the lower error bar is zero for exactly 26 years in the past. I will also assume the line since September 1996 is also exactly 0 as per the following from Nick Stokes’ site:
Temperature Anomaly trend:
Sep 1996 to Oct 2014
Rate: 0.000°C/Century;
CI from -1.106 to 1.106
First of all, I will discuss Lord Monckton’s slope of zero for a time that is slightly larger than 18 years. Lord Monckton says the slope is zero for slightly longer than 18 years as is shown by the flat turquoise line above that starts in September 1996. Another way of saying this is that when we include error bars, there is a 50% chance that cooling occurred during this time and there is a 50% chance that warming occurred during this time.
According to my interpretation of the numbers from Nick Stokes’ site, there is a 95% chance that the real slope for this period of over 18 years is +/- 1.106 degrees C/Century. The two sloping lines from September 1996 show this range. This implies there is a very small chance there is cooling of more than 1.106 C/Century. At the same time, there is the same small chance of warming at more than 1.106 C/Century.
Before I discuss Dr. McKitrick’s 26 years, I would like to offer this quote from Peterson et al., 2009: State of the Climate in 2008, American Meteorological Society Bulletin.
”The simulations rule out (at the 95% level) zero trends for intervals of 15 yr or more, suggesting that an observed absence of warming of this duration is needed to create a discrepancy with the expected present-day warming rate.”
From the above, it appears that climate scientists do not attach a huge amount of importance to the time for a slope of zero, but rather to the time that the warming is not statistically significant at the 95% level.
What Dr. McKitrick has found is that for RSS, the warming is not statistically significant at the 95% level for 26 years. So if WoodForTrees.org (WFT) gives a warming rate of X C/year, the error bars are also +/- X C/year.
According to WFT, there is warming from 26 years ago at the rate of 0.0123944 C/year. So this means that we can be 95% sure the real warming rate is 0.0123944 C/year +/- 0.0123944 C/year. Doing the adding and subtracting, this gives, at the 95% level, a range of between 0.0247888 C/year (or 0.025 C/year to two significant digits) and zero. These two ranges are indicated on the graph above starting at November 1988. Since the lower number is zero and therefore not positive, it is reasonable to say the warming since November 1988 is not statistically significant, at least according to RSS.
Analogous to the case with no warming, there is a small chance that the warming over 26 years is larger than 0.025 C/year. However there is the same small chance that there has been cooling over the last 26 years according to Dr. McKitrick’s calculations using the RSS data.
In the sections below, as in previous posts, we will present you with the latest facts. The information will be presented in three sections and an appendix. The first section will show for how long there has been no warming on some data sets. At the moment, only the satellite data have flat periods of longer than a year. The second section will show for how long there has been no statistically significant warming on several data sets. The third section will show how 2014 to date compares with 2013 and the warmest years and months on record so far. The appendix will illustrate sections 1 and 2 in a different way. Graphs and a table will be used to illustrate the data.
Section 1
This analysis uses the latest month for which data is available on WoodForTrees.com (WFT). All of the data on WFT is also available at the specific sources as outlined below. We start with the present date and go to the furthest month in the past where the slope is a least slightly negative. So if the slope from September is 4 x 10^-4 but it is – 4 x 10^-4 from October, we give the time from October so no one can accuse us of being less than honest if we say the slope is flat from a certain month.
1. For GISS, the slope is not flat for any period that is worth mentioning.
2. For Hadcrut4, the slope is not flat for any period that is worth mentioning. Note that WFT has not updated Hadcrut4 since July and it is only Hadcrut4.2 that is shown.
3. For Hadsst3, the slope is not flat for any period that is worth mentioning.
4. For UAH, the slope is flat since January 2005 or 9 years, 10 months. (goes to October using version 5.5)
5. For RSS, the slope is flat since October 1, 1996 or 18 years, 1 month (goes to October 31).
The next graph shows just the lines to illustrate the above. Think of it as a sideways bar graph where the lengths of the lines indicate the relative times where the slope is 0. In addition, the upward sloping blue line at the top indicates that CO2 has steadily increased over this period.
When two things are plotted as I have done, the left only shows a temperature anomaly.
The actual numbers are meaningless since the two slopes are essentially zero. No numbers are given for CO2. Some have asked that the log of the concentration of CO2 be plotted. However WFT does not give this option. The upward sloping CO2 line only shows that while CO2 has been going up over the last 18 years, the temperatures have been flat for varying periods on the two sets.
Section 2
For this analysis, data was retrieved from Nick Stokes’ Trendviewer available on his website <a href=”http://moyhu.blogspot.com.au/p/temperature-trend-viewer.html”. This analysis indicates for how long there has not been statistically significant warming according to Nick’s criteria. Data go to their latest update for each set. In every case, note that the lower error bar is negative so a slope of 0 cannot be ruled out from the month indicated.
On several different data sets, there has been no statistically significant warming for between 14 and almost 22 years according to Nick’s criteria. Cl stands for the confidence limits at the 95% level.
Dr. Ross McKitrick has also commented on these parts and has slightly different numbers for the three data sets that he analyzed. I will also give his times.
The details for several sets are below.
For UAH: Since June 1996: CI from -0.037 to 2.244
(Dr. McKitrick says the warming is not significant for 16 years on UAH.)
For RSS: Since December 1992: CI from -0.018 to 1.774
(Dr. McKitrick says the warming is not significant for 26 years on RSS.)
For Hadcrut4.3: Since April 1997: CI from -0.010 to 1.154
(Dr. McKitrick said the warming was not significant for 19 years on Hadcrut4.2 going to April. Hadcrut4.3 would be slightly shorter however I do not know what difference it would make to the nearest year.)
For Hadsst3: Since December 1994: CI from -0.007 to 1.723
For GISS: Since February 2000: CI from -0.043 to 1.336
Note that all of the above times, regardless of the source, with the exception of GISS are larger than 15 years which NOAA deemed necessary to “create a discrepancy with the expected present-day warming rate”.
Section 3
This section shows data about 2014 and other information in the form of a table. The table shows the five data sources along the top and other places so they should be visible at all times. The sources are UAH, RSS, Hadcrut4, Hadsst3, and GISS.
Down the column, are the following:
1. 13ra: This is the final ranking for 2013 on each data set.
2. 13a: Here I give the average anomaly for 2013.
3. year: This indicates the warmest year on record so far for that particular data set. Note that two of the data sets have 2010 as the warmest year and three have 1998 as the warmest year.
4. ano: This is the average of the monthly anomalies of the warmest year just above.
5. mon: This is the month where that particular data set showed the highest anomaly. The months are identified by the first three letters of the month and the last two numbers of the year. Note that this does not yet include records set so far in 2014 such as Hadsst3 in June, etc.
6. ano: This is the anomaly of the month just above.
7. y/m: This is the longest period of time where the slope is not positive given in years/months. So 16/2 means that for 16 years and 2 months the slope is essentially 0. Periods of under a year are not counted and are shown as “0”.
8. sig: This the first month for which warming is not statistically significant according to Nick’s criteria. The first three letters of the month are followed by the last two numbers of the year.
9. sy/m: This is the years and months for row 8. Depending on when the update was last done, the months may be off by one month.
10. McK: These are Dr. Ross McKitrick’s number of years for three of the data sets.
11. Jan: This is the January 2014 anomaly for that particular data set.
12. Feb: This is the February 2014 anomaly for that particular data set, etc.
21. ave: This is the average anomaly of all months to date taken by adding all numbers and dividing by the number of months.
22. rnk: This is the rank that each particular data set would have if the anomaly above were to remain that way for the rest of the year. It may not, but think of it as an update 50 minutes into a game. Due to different base periods, the rank is more meaningful than the average anomaly.
| Source | UAH | RSS | Had4 | Sst3 | GISS |
|---|---|---|---|---|---|
| 1.13ra | 7th | 10th | 9th | 6th | 7th |
| 2.13a | 0.197 | 0.218 | 0.492 | 0.376 | 0.59 |
| 3.year | 1998 | 1998 | 2010 | 1998 | 2010 |
| 4.ano | 0.419 | 0.55 | 0.555 | 0.416 | 0.66 |
| 5.mon | Apr98 | Apr98 | Jan07 | Jul98 | Jan07 |
| 6.ano | 0.662 | 0.857 | 0.835 | 0.526 | 0.92 |
| 7.y/m | 9/10 | 18/1 | 0 | 0 | 0 |
| 8.sig | Jun96 | Dec92 | Apr97 | Dec94 | Feb00 |
| 9.sy/m | 18/5 | 21/11 | 17/7 | 19/11 | 14/9 |
| 10.McK | 16 | 26 | 19 | ||
| Source | UAH | RSS | Had4 | Sst3 | GISS |
| 11.Jan | 0.236 | 0.261 | 0.508 | 0.342 | 0.68 |
| 12.Feb | 0.127 | 0.161 | 0.305 | 0.314 | 0.43 |
| 13.Mar | 0.137 | 0.213 | 0.548 | 0.347 | 0.70 |
| 14.Apr | 0.184 | 0.251 | 0.658 | 0.478 | 0.71 |
| 15.May | 0.275 | 0.286 | 0.596 | 0.477 | 0.78 |
| 16.Jun | 0.279 | 0.346 | 0.620 | 0.563 | 0.61 |
| 17.Jul | 0.221 | 0.351 | 0.543 | 0.551 | 0.52 |
| 18.Aug | 0.117 | 0.193 | 0.669 | 0.644 | 0.69 |
| 19.Sep | 0.186 | 0.206 | 0.593 | 0.574 | 0.76 |
| 20.Oct | 0.243 | 0.272 | 0.613 | 0.529 | 0.76 |
| Source | UAH | RSS | Had4 | Sst3 | GISS |
| 21.ave | 0.201 | 0.254 | 0.565 | 0.482 | 0.66 |
| 22.rnk | 7th | 7th | 1st | 1st | 1st |
If you wish to verify all of the latest anomalies, go to the following:
For UAH, version 5.5 was used since that is what WFT uses.
http://vortex.nsstc.uah.edu/public/msu/t2lt/tltglhmam_5.5.txt
For RSS, see: ftp://ftp.ssmi.com/msu/monthly_time_series/rss_monthly_msu_amsu_channel_tlt_anomalies_land_and_ocean_v03_3.txt
For Hadcrut4, see: http://www.metoffice.gov.uk/hadobs/hadcrut4/data/current/time_series/HadCRUT.4.3.0.0.monthly_ns_avg.txt
For Hadsst3, see: http://www.cru.uea.ac.uk/cru/data/temperature/HadSST3-gl.dat
For GISS, see:
http://data.giss.nasa.gov/gistemp/tabledata_v3/GLB.Ts+dSST.txt
To see all points since January 2014 in the form of a graph, see the WFT graph below. Note that Hadcrut4 is the old version that has been discontinued. WFT does not show Hadcrut4.3 yet.
As you can see, all lines have been offset so they all start at the same place in January 2014. This makes it easy to compare January 2014 with the latest anomaly.
Appendix
In this part, we are summarizing data for each set separately.
RSS
The slope is flat since October 1, 1996 or 18 years, 1 month. (goes to October 31)
For RSS: There is no statistically significant warming since December 1992: CI from -0.018 to 1.774.
The RSS average anomaly so far for 2014 is 0.254. This would rank it as 7th place if it stayed this way. 1998 was the warmest at 0.55. The highest ever monthly anomaly was in April of 1998 when it reached 0.857. The anomaly in 2013 was 0.218 and it is ranked 10th.
UAH
The slope is flat since January 2005 or 9 years, 10 months. (goes to October using version 5.5 according to WFT)
For UAH: There is no statistically significant warming since June 1996: CI from -0.037 to 2.244. (This is using version 5.6 according to Nick’s program.)
The UAH average anomaly so far for 2014 is 0.201. This would rank it as 7th place if it stayed this way. 1998 was the warmest at 0.419. The highest ever monthly anomaly was in April of 1998 when it reached 0.662. The anomaly in 2013 was 0.197 and it is ranked 7th.
Hadcrut4.3
The slope is not flat for any period that is worth mentioning.
For Hadcrut4: There is no statistically significant warming since April 1997: CI from -0.010 to 1.154.
The Hadcrut4 average anomaly so far for 2014 is 0.565. This would rank it as 1st place if it stayed this way. 2010 was the warmest at 0.555. The highest ever monthly anomaly was in January of 2007 when it reached 0.835. The anomaly in 2013 was 0.492 and it is ranked 9th.
HADSST3
For HADSST3, the slope is not flat for any period that is worth mentioning. For HADSST3: There is no statistically significant warming since December 1994: CI from -0.007 to 1.723. The HADSST3 average anomaly so far for 2014 is 0.482. A new record is guaranteed. 1998 was the warmest at 0.416 prior to 2014. The highest ever monthly anomaly was in July of 1998 when it reached 0.526. This is also prior to 2014. The anomaly in 2013 was 0.376 and it is ranked 6th.
GISS
The slope is not flat for any period that is worth mentioning.
For GISS: There is no statistically significant warming since February 2000: CI from -0.043 to 1.336.
The GISS average anomaly so far for 2014 is 0.66(4). This would rank it as first place if it stayed this way. 2010 was the warmest previously at 0.66(1). The highest ever monthly anomaly was in January of 2007 when it reached 0.92. The anomaly in 2013 was 0.59 and it is ranked 7th.
Conclusion
There are different ways of deciding whether or not we are in a pause. We could say that if we have a flat slope for X number of years, we are in a pause. Or we could say that if the warming is not statistically significant for over 15 years, we are in a pause. Or we could say that as long as the satellite data sets do not break the 1998 record, we are still in a pause.
In my opinion, a combination of UAH and RSS needs to show statistically significant warming for less than 15 years before I am comfortable with declaring the pause over. What do you think?
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
I think that I don’t care whether it gets warmer or cooler as that proves nothing about CO2. It’s quite easy to prove that CO2 is not a danger in regards to temperatures and though the long pause in temperatures has been useful in making people question Al Gore and other con artists, it might not last. I would rather focus all my energy on educating people on why CO2 can’t, under any circumstances, ever be responsible for dangerous rises in temperature, than spend my time going “look, it’s still cold!” As there is a 33.3333% chance it might get hotter again! (33.3333% it will get colder, 33.3333% it will remain the same)
I agree.
We don’t even need a statistical analysis.
We know that CO2 increased linearly since 1958. Not so for temperature. This is known.
As far as the temperature graphs/analysis presented here they are not informative to convince Al Gore or anybody else.
What we need is simple:
Plot the data points, run a linear regression analysis and plot the line, give the p value and r squared value.
Easy to do:
http://blog.minitab.com/blog/adventures-in-statistics/how-to-interpret-a-regression-model-with-low-r-squared-and-low-p-values
After looking at the results adjustment for autocorrelation can be undertaken if warranted.
Problem is most people cannot relate to that kind of discussion. However, they can relate to the concept of temperature not going up (especially if it is going down). For that reason alone this topic is probably the most important one for winning the minds of the common man.
Your graphical representation of the confidence intervals for trends is incorrect. You have the three lines (the regression line and the upper and lower CIs for trend) meeting at the start of the regression line. In fact, they should meet in the middle of the date range, like this:
http://www.woodfortrees.org/plot/rss/last:312/plot/rss/last:312/trend/plot/rss/last:312/trend/detrend:0.3222544/offset:0.161127/plot/rss/last:312/trend/detrend:-0.3222544/offset:-.161127
Thank you!
I think the point is just to show what trends at the end of the CI range look like. For that, it doesn’t matter what offset you use, and midpoint matching is not really better. A plot that treated intercept correctly would show limit curves looking like hyperbolae. But you can’t do that with WFT.
With all the extra ice for the last 2 years it is hard to see the land sea adjusted data as correct. Cannot change the thermometers so surely at some time the thermometers must move down in accord with the satellite data. The C and W style rigging does allow Polar data to be rigged up and is out of satellite range at times. Is this where the major discrepency is?
The near El Niño meant a warmer year in the tropics, but did it also mean the warm water did not go pole wards.
According to Bob Tisdale, it was the north east part of the Pacific that had an extremely high anomaly lately. But for some reason, this warm water here did not affect the satellite data as it did the surface data. Perhaps less evaporation occurred due to lower absolute temperatures so there was less condensation higher up.
An appeal again:
Does anyone have a graph of the actual global temp (not the anomaly, so around 14.5c) for say the past 100 years…IN DEGREES C? Many thanks all.
Jim,
I have a chart that dbstealy produced some time ago that does exactly that. /Users/paulbhull/Documents/Science/AGW Answers/dbstealeyco2vst.pdf
pbh
Well that didn’t work so well. I could email it, but that would require and email address. Happy to send it if we can bridge the email divide.
pbh
Ghost,
Is this the one?
If you wish to verify all of the latest anomalies, go to the following:
Thanks for including the latest links (-:
Sometimes I have just got to laugh out loud when I read the latest from the statistical torture chamber!. Nothing personal against anyone who likes to join the farce, we all have to get our jollies somewhere. I am sure that you are just like me, Mr Brozek, you were there when the temperatures were being taken. Obviously not during the stone age when the cavemen used to rush outside and dip their finger into an elk carcase then note down the temperature to the nearest thou and paint it on the cave in Woad. Or even when the Roman soldiers used to stop on the way to slaughter a few more Gauls (and what is not to like about that!) and carried their Stephenson screen up the slope and then wrote the results on a slate to the nearest thou. No, I was there in the 60’s when recording data was done on a thermometer to the nearest half degree. I operated a mobile maritime weather station with a mark 1 eyeball and a mark 2 plastic bucket for sea temperature. I used to send the Cadet up to the monkey island to read the temps in a Stephenson screen bolted a few feet above a red painted steel deck. Other ships had green paint, others bitumen tar and rubber. We threw a bucket over and hauled up a bit of seawater. Sometimes to the main deck, sometimes to the bridge wing , sometimes in very rough weather a phone call was made to the ER to get the intake sea temperature. There were no Argo Buoys, no satellites, no digital readouts, precious few aircraft overhead and almost no ships south of 60 or north of 60 degrees. All this was done so that we were not going to get caught out by undetected Hurricanes and Storms. The data was resolved on a slide rule and graphed on a chart with a calligraphy pen then, if we were lucky, faxed back to us as a coherent whole.
I am very sorry if I refuse to get excited about computer programs that forecast the end of the world by torturing data til it screams. Get back to me in a thousand years when we have enough reliable data to even BEGIN to draw conclusions from it.
Sorry for a bit OT and perhaps a dumb question. I have seen the graph at top and ones like it with 1998 way up there hundreds of times. I’m assuming from other data sets as well.
What I can’t reconcile is the recent headlines by NOAA that 2014 may be a record warm year. Looking at the above chart it is not even close. Specifically the press releases have pointed out October 2014 surpassing 1998 by .04 C.
I just don’t see how they are even close.
The above is for RSS only. However UAH also shows 1998 way up there and on both satellite data sets, there is no way that either will come in first or even second. But the surface data sets do indeed show a record to this point. See row 22 of the table where I give the present rankings after 10 months. RSS and UAH version 5.5 are in 7th place, but Hadcrut4.3, GISS and Hadsst3 are all in first place. Hadcrut4.3 and GISS could still end up as first, second or third, but Hadsst3 is guaranteed to set a new record in 2014.
thank you. I get it now. appreciate it.
At the end of the day it’s rather pointless. The Chicken Little’s will simply find some other crisis to blame Carbon (CO2) for, they always do. Because everything evil under the sun is a direct result of Carbon emissions, you see. If you made a list off all things it has been blamed for, it would be as large as Manhattan island (which will soon be underwater BTW).
And even if everyone of these prophecies turns out to be incorrect (which is likely) the Chicken Little’s will still think it is better to have done something. Because you can’t tilt at windmills until you build them first.
There is a big problem with their logic. It is one thing to say that warming will happen due to the greenhouse effect. So if warming occurs, oceans will expand, etc. But if warming is not occurring, then by what mechanism does extra CO2 cause more tornadoes for example?
Good article Werner well written!
As an aside, are there any Brits out there who were listening/watching the breakfast news prog at around 7am this morning? I was munching thru my cereal & slurping my tea, when I heard something about Climate Change being “partly” caused by Human activity! Was I hearing things? Did anyone else hear it? Is this the BBC hedging its bets or something? Is there a subtle sea change going on & I have missed it?
I wasn’t paying perfect attention to Roger Harrabin but I thought he was talking about 2014 being the hottest year since the start of the CET centuries ago and that that was “partly” caused by human activity.
That is not quite the same.
Thanks, that clears that up, I wasn’t paying much attention to RH’s speil that’s why I was rather surprised! It seems from the posts here that NOAA, The Wet Office (UK), & the BBC are bigging up 2014 as the hottest evvaaa!!!
Probably true.
Or there are lies, damm lies and Statistics
Some people use statistics like a drunk uses a lamppost: more for support than illumination.
Now that the “Pause” is over, how will the Team remove it as they did the MWP?
OT but moderator please make this exception and accommodate me.
Jean Beliveau, beloved Captain of the legendary Montreal Canadiens died this week at 83. Beliveau was a great hockey player and a true gentlemen, respected by all.
As a kid growing up in small town Quebec, I lived for les Habs, and look back on their golden years with wonder and gratitude. Can you imagine being a young hockey fan and having your team win ten Stanley Cups, including five in a row? Only the New York Yankees of the same era had as great a winning record!
Our local Wolf Cubs and Boy Scouts held a father-and-son dinner and Jean Beliveau and Gump Worsley (Goalie for the New York Rangers and later for the Canadiens) were our guests of honor. Beliveau had recently joined the Canadians and was an immediate star (we did not use the word “superstar” in those days).
Worsley, who lived in our town, arrived on time but Beliveau was very late so we stated to eat our dinners. At one point, I had a sudden thought and excused myself – I was going to find Jean Beliveau. The school hallways were dark and lit only by red exit lights, and I went toward the Principal’s office in the oldest part of the school, turned a corner and there in the darkness stood a tall man, who asked me “Where’s da gym”? I said “I’ll take you there”, and entered our gym beside Jean Beliveau, as the crowd erupted into cheers and applause.
As I grow older, I am much less interested in sports. Ironically, when I go to my gym I watch more and more sports and less and less news, because I find the news saddens me – the news is really the bad news – it gathers up all the misery and inhumanity of the world and brings it into your living room. In sports, on the other hand, everybody lives to play another day.
Strive to be kind to one another.
Best wishes to all, Allan
Michael Farber on Jean Beliveau:
http://www.tsn.ca/video/beliveau-a-legend-on-and-off-the-ice-1.151132
As an ex Montrealer of a certain age, I remember Beliveau very fondly. He was their leader in the glory years. Rarely do you find an athlete who has the appearance, dignity and gravitas of a ‘statesman’ on and off the ice.
He was thoroughly respected by one and all – a class act.
The Montreal Canadiens are one of the best franchises in pro sports for recognizing their history and former stars – the tradition continues!
The good Lord’s slope is zero because that is his goal. His point is not that the slope is zero, his point is that you have to have greater than N years of data to have a non-zero slope. Different meaning to the various sets of data.
From the article:
There are different ways of deciding whether or not we are in a pause… In my opinion, a combination of UAH and RSS needs to show statistically significant warming for less than 15 years before I am comfortable with declaring the pause over.
This is ridiculous, IMHO. People on both sides of the debate are splitting hairs over hundreths of a degree. If global warming was a problem, we would know it.
For some needed perspective:
http://suyts.files.wordpress.com/2013/02/image266.png
An odd idea of perspective! Arbitrary scale from 0 to 120 (Fahrenheit I assume?).
Actual global average temperatures on this planet have not strayed more than about 25F above current levels and 12F below current levels within the last 500 million years.
The difference between full on glaciation and the mildest interglacial period is only about 10F. So what on earth does a 120F range have to do with anything?
Why not show a range from absolute zero to the surface temperature of the sun?
I believe dbstealey’s graph shows what the global temp trend looks like on a home thermometer.
Nigel Harris,
The scale is not important. What matters is the fact that by using an x-axis scaled in degrees, there is no observable change in T. The original argument claimed runaway global warming and predicted it would accelerate. If there was runaway warming occurring, it would be easily visible in that chart.
But the original conjecture was wrong. Everything said subsequently is backing and filling: trying to convince people that there is a problem, when there isn’t. It’s just crying “Wolf!!” The public is getting jaded. Can you blame them?
Also, there are other charts showing the same thing. All we are observing are normal step change rises from the depths of the LIA. There is nothing either unprecedented or unusual happening. Everything we see now has happened before, and prior to human CO2 emissions being a factor.
==============================
Dave in Canmore,
Yes, that’s what it shows: no trend. I worked in a Metrology lab for thirty years, calibrating weather instruments including every kind of thermometer, both stick and electronic; PRT, thermocouple, RTD, etc. Anyone who tells you that you can accurately measure tenths and hundreths of a degree C/F accurately, without using *very* expensive instruments, simply does not know what they’re talking about. Changes of ±0.1ºC are bandied about as if that is reality. It isn’t. Almost all state of the art thermometers have larger error bands. In most cases, much larger.
I prefer using degrees C or F. If a problem is brewing, those will show it just as fast as any electronic thermometer. The problem for warmists is that they can’t show any kind of a problem. Everything currently observed has happened before, repeatedly, and to a much greater degree.
For all practical purposes, global warming has stopped, for many years now. It may stay the same, or resume, or cooling may begin. We don’t know. The only thing that has not stopped is the constant climate alarmism, by people who have made consistently wrong predictions from the very beginning.
When I look at that chart I see zero chance in hell of the predicted disasters of CAGW manifesting. It is certainly non threatening, although I am a bit curious, as I thought we had warmed 3 to 4 F since the little ice age, and I do not see it in the F thermometer chart. (I say the further from an ice age the better, within reason)
0.8 C or 1.4 F since 1880 seems about right.
dbstealey, I can still see some wiggles in your plot. Convert to Kelvins to really flatten that sucker out.
No one claims there are no wiggles. They are just small and inconsequential.
Werner,
Especially when playing visual games with the y-axis. If I want to exaggerate a warming trend by printing out a plot on legal size paper with the y-axis going lengthwise it will have dramatic visual impact. But the significance of the change is due to the physics of the actual phenomena, not the big scary units I choose. Not whether I express the figures as absolute values on that scale or as anomalies. The planet does not read charts. It is not fooled by appeals to (in)credulity by use of small or large numbers.
One way to cut through all these stupid and scientifically useless rhetorical games with smallish temperature values vs. 10^infinity energy statistics is to strip units out of it entirely and think in ratios or percentages. My favorite example is to note that from bottom to top of the temperature cycle from glacial to interglacial is ~12 K. Since 1880, global temperatures have changed on the order of 0.8-0.9 K, near enough to one to call that 1/12th of a full glacial cycle. The dreaded 2 K anomaly is 1/6th of same. Considering that bottom to top of the natural cycle takes on the order of 20,000 years and we’ve just experienced 1/12th of that in 135 years … well you can do the math, it’s 150 times the maximum average observed rate over the past million years when the planet is left to its own devices.
No shenanigans with charts, no hard choices with units since temperature and energy have a one to one relationship. I don’t care what planet you’re from, 1/12th a movement in 0.007th the time is nowhere remotely close to insignificant.
But the significance of the change is due to the physics of the actual phenomena, not the big scary units I choose.
I agree! And the people that really need to know this is those who insist on giving steep rises to ocean heat content by showing X times 10^22 Joules on the y axis. Then when you convert to degrees C, you find it is about 0.1 C that the ocean went up in the last 60 years. So if you wish to multiply both numbers by 12, you get 1.2 C in 720 years. Sorry, but it does not alarm me if the ocean goes from 3.0 C to 4.2 C in 720 years. For all intents and purposes, the ocean is an infinite heat sink for minor warming of the air.
Werner,
Well gee, Joules are the SI standard unit of energy. Poor science to use appropriate units, yes indeed.
And now we’re back to playing games with small numbers. Reminds me of this rather silly (read: dishonest) chart from Bob Tisdale:
10^22 Joules and °C on the same axis!!! Switch to 10^24 Joules and voila, you can actually derive some meaning from the plot, plus the y-axis numbers aren’t quite as big and scary either.
Totally, utterly non-scientific tripe. The relative slopes don’t mean diddly squat if you’re just arbitrarily scaling stuff for purposes of visual presentation, especially since there’s a direct linear relationship between temperature and energy for crying out loud.
Over a 100 ky glacial cycle, it works out that surface temps change by a factor of about 5 times greater than the deep ocean. The lag is about 10k years for each major reversal in surface temps. But that’s deep ocean now. The surface — where all the ice hangs out — is far more responsive. Effective heat sink, yes. Infinite, not so much. All depends on one’s own personal expiration date, I suppose.
And now we’re back to playing games with small numbers. Reminds me of this rather silly (read: dishonest) chart from Bob Tisdale:
I do not agree that this is dishonest at all. Suppose you went swimming in the ocean and you found it rather chilly. And then you were told that 10^23 joules had been added to the water since last year. Of course joules are the SI unit for energy, but would that mean anything to you? You might get the impression the oceans were about to boil. But if you were then told the temperature went up by 1 C, that would be much more meaningful. As well, you might not even be able to detect that difference.
I believe that we will either have an ice age or nuclear fusion long before our world has over heated.
Werner,
Yes. What I didn’t get from high school I got in my freshman year at college. It’s not the researchers’ fault that some people forgot their standard education as soon as they finished their exams.
There’s another way to slice through all this b/s, and that’s to use Watts per sq. meter. A per unit area calculation makes things easily comparable because it strips out the vast differences in total mass and relative heat capacity allowing apples to apples comparisons instead of donkeys to elephants.
My point was that when I get up in the morning and need to decide whether to wear long johns or not, I check the TV for the temperature. I am not interested in how may joules or W/m2 the air over my city gained or lost since it was -5 C last night.
The problem is not that I could not make the calculations to convert joules to change in C using Q = mct, but that would not be nearly as convenient as seeing the temperature.
Werner,
I agree, and for the vast majority of us (including me) that does hold true. My point is twofold:
1) When we go skinny dipping in the ocean, we don’t dive to 2,000 m.
2) Averaged temperature across a 2 km thick layer of sea water cited in exclusion of all else obscures what’s going on at the top most layer where we … and things like ice … live.
Pile on to (2): plotting °C and 10^22 J down to 2,000 m on the same axis is worse than meaningless when the obvious intent is to hide the incline of °C. I could only laugh at the sheer audacity of it.
Again I agree, but my main points are not about that, but rather the scientific, physical, relevance of a given choice of units. When possible I like to think in Watts per sq. meter, but of course that only makes sense at boundary layers. OHC in Joules is the most direct way to that calculation. When talking °C, well, yes of course ∆T is an impressively small number down to 2,000 m. Not so small higher up the water column: https://drive.google.com/file/d/0B1C2T0pQeiaSdmJUcmpJQkVCWVE/view?usp=sharing
The rate plot at the bottom is fun, isn’t it.
Thank you! I still prefer temperature and the differences among all layers is fine as you have shown. And while the top 100 m is warming fastest, I see nothing alarming about it. And should the rate of warming at the top increase, it will just go to the lower layers faster and the warming at the top would be dampened.
Werner Brozek,
Emotional reaction isn’t always a choice. Opinion often is. Truth be told, I don’t worry much about estimates of the worst effects because I’ll be dead. Aside from that, I don’t think were in danger of extincting ourselves by CO2. Personally I think it’s more likely that Pakistan will nuke India over Kashmir, or some other similar scenario.
The rate plot shows all three depth layers with an accelerating warming trend across the entire record. [1] Yes of course the cooler depths dampen the warming at the surface, the scientifically relevant question at this point is: how much, and what will the energy which remains at the surface do in the future. And where, as in what surface grid. Three curves on a plot of global averages doesn’t get remotely close to conveying that sort of info. Does your “I see nothing to be alarmed about” statement contain the barest of hint of research or calculation into local surface effects?
———————————
[1] The surface rates are the outlier here, showing deceleration. All three trendlines are incredibly sensitive to start and endpoint, but especially at the surface. If I knock out the first two years of the record, the surface acceleration trend goes significantly more positive than the three depth curves, which I would of course, expect. This is why I’m dubious of trend analysis as a primary means of investigation and prediction.
O.K. Let us take any emotion out of it. A slope line was not shown, but the top 100 m were at 0.10 in 1978 and 0.33 in 2013, which is an increase of 0.23 in 35 years. This is way less that 1.0 C in 100 years. And an extra 1.0 C of warmer ocean will only warm air by 1.0 C as well. And if an extra 2 C above 1750 is supposed to be bad, we have a long time before we reach that point. I am sure technology advances over the next 100 years will allow us to cope with what needs to happen.
Furthermore, I do not believe that an extra 2 C will be that bad. Where I live in Canada, I would still spend a lot of money keeping warm and none keeping cool.
Werner,
Your eyeballs don’t deceive. 0.241 °C change, rate 0.007 °C/year.
0.7 °C/century, 30% shy of a full degree is significantly less. But you assume rate will stay constant. I did throw linear trendlines on the bottom graph, and the 100 m curve shows an acceleration of 0.000357 °C/year^2. The predicted rate from that regression in 2014 is 0.0135 °C/year, and the predicted temp is 0.35 °C, (+0.02 of the actual).
Now if I take my turn to assume, at constant acceleration the 100 m temp works out to 2.84 °C by 2100. That’s the anomaly above the 1955-1964 baseline average, not from 1750. The rate of change in 2100 would be 0.0441 °C/year.
Oh probably. I’ve said before we’re tenacious and creative. I don’t think we’ll extinct ourselves. But I think you’re getting ahead of things here; the data show an accelerating rate of temperature change, not a constant one.
[chortle] You could always take the reverse of markx’s philosophy to heart right now and move to Florida: http://wattsupwiththat.com/2014/12/05/friday-funny-over-a-centurys-worth-of-failed-eco-climate-quotes-and-disinformation/#comment-1807203
Thank you for that! But who knows if the rate will accelerate until 2100? Climate seems to go in 60 years cycles in addition to other cycles. And even if it does continue to accelerate, people will have to adapt to whatever circumstances they find themselves in.
The worst thing our Alberta government can do now is to spend 2 billion on carbon capture to possibly shave off 1/10000 of a degree by 2100.
The RSS data also shows a (slightly) negative trend from Jan 1979 to Sep 1989 or 10 years, 9 months. That means that all of the warming is confined to a period of less than seven years.
This would be further proof that CO2 is not a major player.
Sigh, a linear fit to a nonlinear process yet again. If you look at the satellite record from 1980 to now you will see that it is a step function with a flat region prior to 1998 and then a step up to the regime we are in now. Both of the steps have fluctuations that are due to El Nino/La Nina. At 1998 there is a huge El Nino (+ something else?) and then the new plateau. Even Trenberth admits this according to Bob Tisdale. The simple minded linear increase due to CO2 predicted by the models cannot produce this kind of behvaior, it can only come out of a nonlinear process, aka, chaos.
Please stop with the silly fits of straight lines to nonlinear processes. And while I’m on a rant, please stop with the smoothing, it throws away data.
This site has various curves:
http://www.climate4you.com/GlobalTemperatures.htm
However WFT does not have that function so I use the tools I have. I believe these tools are adequate to indicate the climate models are in trouble.
Unfortunately the tools you are using are making things worse.
Look at the plot of CO2 data from 1958 to now from Hawaii. Do you need any statistical analysis to see that there is a linear increase in CO2 concentration from 1958 to now?
Nobody needs a statistical analysis for this. Just a pair of eyes and a brain.
Now, while CO2 and temperature anomalies were increasing modelers were doing fine.
Claiming that CO2 was the cause of the increase, regardless of the fact that while the CO2 increase was linear, the temperature anomalies were not increasing in a linear fashion. But still, both were increasing.
Now, however, this is no longer the case. While CO2 is still increasing we now have a flat region for temperature anomalies over a number of years. The modelers were doing fine with short flat regions, they could ignore such.
They are now in trouble. The longer the current flat region will stay and the longer the increase in CO2 will continue, the more difficult it will be for them to claim that CO2 is responsible. They now need to adjust their “models” to fit the current data. I don’t know how they will do this. But if they can’t fit historical data, they can’t predict!
The temperature also did not increase from 1958 to around 1977. Nor from about 1944 to 1958, even though CO2 was rising then, too, just not recorded at Mauna Loa. There were just two decades in the middle, ~1977-98 (or ’96), when rising CO2 happened to coincide with apparently rising T, as it no longer is.
An awful lot of discussion (not just this post) about such trivial trends. It should be plainly obvious that the trends for the last 10-18 yrs, whatever, are so small in the sat data as to be insignificant. I don’t count land-stations — way too problematic.
When will many heads of state realize how insignificant the warming is?
The most significant trend over the last 25 years has been the unprecedented pace of increase in institutionalized mendacity and the official acceptability of deceit as a means to policy making.
It’s so bad now that offenders are not disgraced in the slightest for being caught Grubering.
Dishonesty has become an official badge of honor.
Werner Brozek says:
This would be further proof that CO2 is not a major player… When will many heads of state realize how insignificant the warming is?
At current CO2 concentrations, changes in temperature due to CO2 are too small to show up in the data. Even a 25% rise in CO2 would not be enough to show a measureable increase in T. That is why there are no comparable charts showing that changes in CO2 cause subsequent changes in T.
Most of the observed changes in CO2 are caused by changes in T. I am willing to be convinced otherwise, but it will require the same kind of data that I posted here.
==============================
Steve Oregon says:
The most significant trend over the last 25 years has been the unprecedented pace of increase in institutionalized mendacity and the official acceptability of deceit as a means to policy making.
It’s so bad now that offenders are not disgraced in the slightest for being caught Grubering.
Dishonesty has become an official badge of honor.
Repeated for effect.
My head is spinning with all the graphs and numbers. But one thing is clear. Every Damn “study” I’ve seen about the “catastrophic” impact of “the global warming” over the past 20 years is Horse spit. If a study says “cumulative effect” it is probably also useless because the weasel word quotient is pretty high, but if the impact has been during the last couple of decades……
The biggest thing I can’t reconcile in my pea brain is this: if we pick a pristine station, well sited, with records far back into the 19th century, in no case does it show what the composites claim. All single sites are rejected unless sliced, diced, and fully homogenized. So how come we accept a single site for CO2 – especially one so “typical” of the rest of the worlds landmass. This warming stuff remains turtles all the way down, and the floggers are the same snake oil salesmen we once would have ridden out of town on a rail.
Seeing as we have the USCRN with all class 1 locations and a very well thought out setup could we use those? I know that is only land and only USA but given the difference between all the other terrestrial based data sets could we use it to calibrate their accuracy? Heck could we use it to check UAH/RSS readings over the area that the CRN covers?
I will be very interested in how RSS and UAH fix their differences.
Those would be excellent questions to post on Dr. Spencer’s site here:
http://www.drroyspencer.com/2014/12/uah-global-temperature-update-for-nov-2014-0-33-deg-c/
By the way, the monthly posting on WUWT of Roy Spencer’s UAH global lower tropospheric temperature update seems to have been missing the past two months. It was +0.39 in October and +0.33 in November, in case anyone is interested.
I am thinking that the pause will soon be known as the plateau and that we are on the back end of that plateau. Once this El Nino is over the cooling will be evident, even on the molested datasets, and they will need to adjust the last decade of temperatures down to hide the cooling as long as they can.
El Niño?
Statistics-newbie question… what is the definition of “Statistically Significant”? Is it 1 standard deviation above a flatline, 2 standard deviations above a flatline, or what?
Here is “or what” with a clear examples given at this site and easy to understand:
http://blog.minitab.com/blog/adventures-in-statistics/how-to-interpret-a-regression-model-with-low-r-squared-and-low-p-values
Here in this post, they are plotting temperature anomalies data vs year. OK.
So, they want to know if there is a trend. OK
Maybe the trend is upward (warming), this could justify that CO2 maybe the cause since CO2 has been increasing during these years.
Maybe the trend is downward (cooling), obviously against CO2.
Maybe there is no trend, obviously against CO2 also.
So, the simple start is to use linear regression analysis. You have two examples from Minitabs cited above, take a look at them and you will easily see how to interpret such data.
The trend is exactly the same for both examples given, but obviously much more variation in one example than the other.
Look at how they measured “how good the fit is”. They used two values: p value and R squared value.
If the p value is <0.05, statisticians will declare "statistical significance" . However, in regression analysis, R squared value is really what you want. This value will be between 0 and 100% (as in the example above, although this is usually given by other statisticians as between 0 and 1 instead of 0 and 100%). Obviously if it is 100% (or 1) you have a perfect fit. So look at the R squared values for the two examples at Minitabs and you will easily see that when you have large variations, even if the p value is <0.05 the R squared value decreases rapidly. This indicates that other than the single factor you selected to plot against is not the only variable contributing (or causing) to the trend. Obviously if there is no upward or downward trends then there is no causation.
Simple linear regression is just a beginning, but at least it should be done properly as a starting point not only to indicate statistical significance but also to give you assurance of declaring that the modeling is working or what other things you need to look at. A simple plot of temperature vs CO2 would be the thing to do!
RD50,
I ginned this up a while back for some other purpose, 1880-present for GISTemp and UAH: https://drive.google.com/file/d/0B1C2T0pQeiaSczJHSG1oU0p6M2c
Just for you, scatterplot of HADCRUT4 from 1850-present and UAH’s entire record against CO2 doubling from 1850: https://drive.google.com/file/d/0B1C2T0pQeiaSWlBoYkRRTXh3TTA
To Brandon Gates. Very nice and yes r squared is what is needed.
I will keep this for sure. Glad I revisited this post this morning.
Thank you.
To Brandon Gates:
Saved your graphs. Worked fine.
Since you gave me something, maybe you do not have a copy of the first paper published on CO2 and temperature. A classic.
Here is a copy of it:
http://onlinelibrary.wiley.com/doi/10.1002/qj.49706427503/pdf
Old style science writing. I love Fig. 2 and the predictions in Table VI.
I am in agreement on page 14 “The conclusion…….” We can still take in some CO2.
There is an interesting discussion after the References section of the paper.
Thanks again.
rd50
RD50,
You’re welcome for the graphs. I did not have a copy of Callendar’s paper. I agree with you, it is a joy to read and quite prescient … and not just from the perspective of the science. First sentence after the abstract:
If he only knew …
RD50, PS;
I just finished reading the Discussion section. While properly skeptical questions and rebuttals — to be expected in the face of such novel research covering a very large scope — they are eerily familiar. One would hope that after nearly 80 years such basic cautions and objections would have been handled to the satisfaction of all. Instead, in some quarters they are hashed and rehashed as if they’d never been asked at all. This paper is an absolute goldmine for perspective. Again my thanks for referring me to it.
For a regression, it is usually the prob that the data can be explained by a line of 0 slope. To be significant, usually a prob of .05 is the cut off, and that’s about 2 S.E. (not S.D.). The 0.05 is negotiable, depending on how $$ the test is, if people are going to die, etc.
Climate science says something like warming is statistically significant if there is a 95% chance of warming actually happening. This is slightly less than the 95.45% that represents two sigma.
I am not a linear tread person, I normally deal with two samples at a time. Really simple stats.
That means I use Student’s T or Wilcoxon Rank for most of my stuff. I prefer the Rank test because it makes no assumptions, although you get the same answer with the T test … usually.
Anyway, if you Wilcoxon Rank using the first 10 months of this years NOAA series, 2014 data is higher than any other year, although not significantly. For example, 2010 has a prob of about .25 and 1998, 2003 and 2005 come in at prob of about 0.1. But other years, like 1995, 1999, or 2003 have probs <.01, which is highly significant. (I didn't run all the years).
If you like, you can run 2014 against a run of years to lower the S.E., and let's include a warm year for a challenge: 2014 against 2009, 2010 and 2011, you get a prob of 0.02. Pretty significant. Or run 2013 and 2014 against 2003 and 2004 (hot and cold in each pair): that is like 0.07, so not 0.05 but not bad.
So I can't really argue with the L.R. people because I am into L.R.s, but this non-significance business doesn't pass the smell test.
(And if I was into L.R., I would use the Kolmogorov-Smirnov test…)
Werner – First, I really don’t like all those straight lines in your first graph. You have sufficient resolution to show actual temperature trend instead of speculative guesses. Secondly, I prefer UAH satellites to RSS because RSS shows cooling in the twenty-first century and UAH does not. I happen to think that this cooling is an artifact of their new data handling procedure. UAH did not monkey with data handling and shows a straight horizontal line for hiatus. And a straight horizontal line fits both satellite data for the eighties and nineties as I shall show. But since we are talking of a 26 year interval you should include the data for that interval by going back to 1979 when satellites came on line. As it is, your first segment is shortened and shows only two of the five El Nino peaks visible in the satellite record before the beginning of the super El Nino. All five are needed for the analysis. I suggest you use UAH for that. Now lets forget about all the trends you have seen and just do the analysis needed. First thing is to make the actual temperature trend visible. Use a transparent red marker wide enough to cover the bulk of the noise covering the temperature curve. The noise is caused by cloudiness variations and hence has an approximate mean amplitude, with occasional outliers you can ignore. That transparent red band is the best possible way to define global air temperature, but it is not global mean. Global mean can easily be defined for the segment in the eighties and nineties that shows ENSO oscillations. There are five El Nino peaks there, with La Nina valleys in between. ENSO amplitude in the eighties and nineties is approximately 0.5 degrees Celsius. The same period shown in ground-based data has an amplitude of about 0.3 degrees Celsius showing the difference in resolution between the two measurement techniques. Once you have the red band drawn in put a yellow dot at the half way mark between an El Nino peak and its neighboring La Nina valley. These dots mark the locations for global mean temperature. Connecting them shows what happens to global mean temperature. Trying to do it by computer is not a substitute because of systematic errors in computer-generated curves. See figure 15 in my book “What Warming?” for an example. Doing this on the left side of the graph will give a horizontal straight line from 1979 to early 1997. It tells us that throughout this period of ENSO ossilations global mean temperature did not change. This is proof that El Ninos have nothing to do with global warming. But if not El Ninos then what? This regular succession of ENSO oscillations is cut short by the super El Nino of 1998. It rises and falls quickly and on both sides of it there is a La Nina depression as there should be. By analogy with the eighties and nineties there should be another El Nino rising in 1999 when the super El Nino has ended. It looks that way but the temperature keeps rising until it is a third of a degree above that of the previous ENSO oscillation in the eighties and nineties. And what is more, temperature stays at that level for the next seven years instead of coming down as expected. So what happened to ENSO? Well, it does show signs of life when the 2008 La Nina arrives. This really confused Trenberth who had no idea why there was cooling when he expected warming. And as we expect, an El Nino is not far behind that La Nina and appears in 2010. Problem is, all these people hoping for the end of hiatus were expecting another El Nino in 2014 or 2015 to save them but have not gotten nothing yet. They are dreaming that an El Nino will cause global warming but as I pointed out, El Ninos have nothing to do with warming. They are all paired with La Ninas, and the average of the two, not the El Nino peak itself, determines global mean temperature. The over-all picture is thus a continuing hiatus/pause, regardless of what an El Nino may not do. It is highly likely that the hiatus we are in was instigated by the huge amount of warm water carried across the ocean by the super El Nino. That super El Nino and its consequences are not well understood but all the billions of research money they get from Uncle Sam are not available for silly things like trying to understand climate. The super El Nino was followed by that mysterious step warming that raised global temperature by a third of a degree Celsius and then stopped. This is actually the only warming the world has seen since 1979, and it is guaranteed not to be anthropogenic. As I pointed out, there was also another hiatus in the eighties and nineties, and the two don’t line up because of this step warming in between. If you take a 26 year segment of temperature history it is comprised of two horizontal segments, one preceding the arrival of the super El Nino, and one following its departure, separated by a temperature rise of a third of a degree at the beginning of the century. Because of this it is impossible to join them into a single curve. Oh, and one more thing. You have not heard of the hiatus of the eighties and nineties because all three ground-based temperature sources (GISS, NCDC, and HadCRUT) are faking a warming there that does not exist. It used to be called the “late twentieth century warming” and claims were made that it must be human caused because no one knew why it was there. I proved this fakery (see Figure 24) when I wrote my book. I even put a warning about it in the preface, but nothing happened. Their cooperation is proven by the fact that their data were computer processed by an identical procedure that left its footprints on publicly available temperature curves. To me, that is scientific fraud. They are still brazenly raising the slope in the twenty-first century graph, with the absurd result that in their temperature curves the 2010 El Nino peak is now higher than the 1998 super El Nino is. My advice is to not use any ground-based temperature curves if satellite data are available.
My advice is to not use any ground-based temperature curves if satellite data are available.
Thank you for these thoughts. I believe Bob Tisdale would agree with much of it. I am not in a position to judge between RSS and UAH. You like UAH and Lord Monckton likes RSS. I give the statistics for both. And by giving the ground based data as well, the glaring discrepancies become apparent.
Maybe we should do an additional graph showing that there hasn’t been any net warming since the MWP 1200 years ago.