From John Ray's shorter notes
January 18, 2018
Cold Facts on the Globe’s Hottest Years
Annenberg fact checking is often recognizably biased so the presentation below has to be taken with a grain of salt. So let me supply some salt
But first let me congratulate author Vanessa Schipani on a scholarly piece of work. It's nice to have a detailed discussion of the numbers for a change. And she does concede in the end that the "hottest year" talk so beloved of Warmists is pretty meaningless, which is pleasing.
So she says that it is the long-term trend which we have to focus on. So far, so good.
But a trend by itself tells you very little. The interesting question is what causes the trend. On (rubbery) NOAA figures there has been a slight trend over the last 150 years but are the details of that trend favourable to the global warming theory? They are not. So we have to move on to matters that Vanessa does not consider. In particular, was the trend in temperature matched by a trend in CO2 levels? That the two trends do coincide is the essence of the global warming theory.
To examine the question, we have to ask what are our start and finish points of any trend we want to examine. It is an old truth of chartmanship that you can prove almost anything by a judicious selection of start and finish points. Every such decision will have a degree of arbitrariness but some are less arbitrary than othrers.
During my research career I did a lot of factor analysis, generally principal components analysis. I even remember centroid analysis! And you can generally get quite a few factors out of a modern analysis. But how do you decide which factors are likely to be important? A very common procedure is to look for the "natural break" in an ordered series of eigenvalues -- sometimes called a "scree test". And looking at any series of numbers can involve a decision of that nature.
So, in the case of the terrestrial temperature series we can see on a number of occasions such "natural breaks". One of them is, quite simply, the 21st century. The 21st century temperatures bob up and down but display no overall trend. There is NO global warming in the 21st century so the trend up to that time appears to have run its course. It is certainly true that El Nino pushed up temperatures in 2015 and 2016 but El Nino is not a product of anthropogenic global warming and its influence has by now just about petered out, leaving the 2017 temperature very close to the pre El Nino average, which gives us temperature stasis back.
And note that CO2 levels did NOT rise during the El Nino warming event. I monitored the CO2 figures from both Cape Grim and Mauna Loa right from the onset of the warming -- beginning roughly in August 2015. And I noted that the 400ppm peak had been reached BEFORE that warming event and then plateaued during the warming event. There was no rise in CO2 levels accompanying the rise in temperature. So the temperature rise COULD NOT have been caused by a CO2 rise -- because there was no CO2 rise. And it's now in the journals that CO2 levels plateaued in 2015 and 2016.
So El Nino did not merely contribute "part" of the 2015/2016 warming event, it contributed the WHOLE of it. So if we remove the influence of El Nino, we can see that there has been NO anthropogenic global warming for the whole of this century. The levels of CO2 have influenced nothing. Warmist theory is wrong
Sen. James Inhofe misleadingly claimed that the statistics behind the globe’s likely hottest years on record — 2014, 2015 and 2016 — were “meaningless” because the temperature increases were “well within the margin of error.” Taking into account the margins of error, there’s still a long-term warming trend.
Inhofe, a longtime skeptic of human-caused climate change, made his claim Jan. 3 on the Senate floor.
Inhofe, Jan. 3: The Obama administration touted 2014, 2015, and 2016 as the hottest years on record. But the increases are well within the margin of error. In 2016, NOAA said the Earth warmed by 0.04 degrees Celsius, and the British Government pegged it at 0.01 Celsius. However, the margin of error is 0.1 degree, not 0.01. So it is all statistically meaningless and below the doom-and-gloom temperature predictions from all the various models from consensus scientists.
Since Inhofe cites data from the National Oceanic and Atmospheric Administration and the British government, we’ll concentrate on their analyses.
According to NOAA, 2016 was the warmest year on record for the globe since record keeping began in 1880; 2015 ranked the second warmest year and 2014 the third warmest. There are uncertainties in those rankings, however.
As we explained in 2015 when then-President Obama proclaimed 2014 “the planet’s warmest year on record,” such a definitive claim is problematic. For instance, while NOAA found then that 2014 had the highest probability of being the warmest, there remained statistical odds that other years could have held that distinction. But as we explained, scientists are more concerned with long-term trends than any given year.
And 2017 is on track to be another warm year. On Dec. 18, NOAA said 2017 could end up being the third warmest on record, based on data for January to November. NOAA spokesman Brady Philips told us the agency will release information on the year as a whole on Jan. 18.
NOAA ranks years by looking at how much their average temperatures differ from the 20th century average — what scientists call a temperature anomaly.
Based on the agency’s analysis, the average temperature for 2016 was 0.94 degrees Celsius (1.69 degrees Fahrenheit) above the 20th century average of 13.9 C (56.9 F). The margin of error for 2016 was plus-or-minus 0.15 C (0.27 F).
NOAA explains that a margin of error takes into account the “inherent level of uncertainty” that comes with “[e]valuating the temperature of the entire planet.”
The agency adds that the reported temperature anomaly — 0.94 C in the case of 2016 — “is not an exact measurement; instead it is the central — and most likely — value within a range of possible values.”
For example, that range, or margin of error, would be 0.79 C (1.42 F) to 1.09 C (1.96 F) for 2016. Scientists at NOAA are 95 percent certain the temperature anomaly for 2016, or for any given year, will fall within the margin of error.
As Inhofe notes, NOAA scientists found that the average temperature for 2015 was 0.04 C less than 2016’s at 0.90 C (1.62 F) above the 20th century norm. The margin of error for 2015 was plus-or-minus 0.08 C (0.14 F), which means the range for 2015 is between 0.82 C (1.48 F) and 0.98 C (1.76 F).
The difference between 2015 and 2014, however, was wider. The average temperature for 2014 was 0.74 C (1.33 F) above the 20th century mean, or 0.16 C (0.29 F) less than 2015. The range for 2014 is between 0.59 C (1.06 F) and 0.89 C (1.60 F).
So the margins of error for these three years do overlap. When we requested evidence from Inhofe’s office, spokeswoman Leacy Burke sent us links to articles that reiterate the senator’s claim that the temperature increase in 2016 was within a margin of error – meaning, again, that while 2016 is most likely the warmest on record, other years that fall within that margin, including 2015 and 2014, could be the warmest. Still that doesn’t mean the statistics are “meaningless.” Over the long haul, data show an increasing trend, as the chart below shows.
“Overall, the global annual temperature has increased at an average rate of 0.07°C (0.13°F) per decade since 1880 and at an average rate of 0.17°C (0.31°F) per decade since 1970,” says NOAA.
Similar to NOAA, the U.K.’s Met Office, the country’s national weather service, reported that 2016 “was one of the warmest two years on record, nominally exceeding the record temperature of 2015.” The agency also found that 2014 likely ranked the third warmest year.
Both NOAA and the Met Office note that human-caused global warming isn’t the only force behind the record temperatures.
Peter Stott, then the acting director of the Met Office Hadley Centre, said: “A particularly strong El Niņo event contributed about 0.2C to the annual average for 2016, which was about 1.1C above the long term average from 1850 to 1900.” El Niņo is a naturally occurring interaction between the atmosphere and ocean that is linked to periodic warming.
Stott added, “However, the main contributor to warming over the last 150 years is human influence on climate from increasing greenhouse gases in the atmosphere.”
The Met Office’s numbers differ slightly from NOAA’s, in part, because the agency uses a different reference point.
NOAA ranks years based on how much their average temperatures differ from the 20th century norm. The Met Office uses the temperature average between 1850 and 1900 or between 1961 and 1990.
Using that latter reference point, the Met Office found 2016’s temperature anomaly to be 0.77 C, plus-or-minus 0.1 C, which was only 0.01 more than 2015’s temperature anomaly.
So Inhofe is right that the British government’s margins of error for 2016 and 2015 overlap.
But Grahame Madge, a spokesman for the Met Office, explained in an email to us why it’s important to look at the long-term trend — not just the difference between two years, as Inhofe did.
Madge, Jan. 6: When looking at global temperature rise it helps to look at the way the stats and figures are framed. For example, 2016 was the warmest year since pre-industrial times. However, it was only marginally warmer than the previous year, which was also a record. When viewed as parallel years, however, they really stand out in the long-term record. … We try to focus on the long term when presenting information. You can make a desert seem like a lush wetland if you only show the oasis.
NOAA also explains the difference between looking at single years versus the long-term trend: “As more and more data builds a long-term series, there is less and less influence of single ‘outliers’ on the overall trend, making the long-term trend even more certain than the individual points along it.”
In other words, if scientists found that the globe had just one year with an exceptionally high temperature average, they may not be convinced that global warming is occurring. But if data show that the planet has experienced a number of record warm years in a row, it suggests the warming trend is real.
In fact, NOAA says there’s only a 0.0125 percent chance of seeing three outliers in a row — and the Earth has seen many more record warm years than three.
NOAA writes that 2016 “marks the fifth time in the 21st century a new record high annual temperature has been set (along with 2005, 2010, 2014, and 2015) and also marks the 40th consecutive year (since 1977) that the annual temperature has been above the 20th century average.”
So while Inhofe was right that the margins of error for temperature measurements in recent years overlap, that doesn’t negate a long-term warming trend or render the temperature anomalies “meaningless.”
Go to John Ray's Main academic menu
Go to Menu of longer writings
Go to John Ray's basic home page
Go to John Ray's pictorial Home Page
Go to Selected pictures from John Ray's blogs