By Christopher Monckton of Brenchley
Now that the small La Niña that has recently ended has begun to have its effect on global temperatures, the UAH monthly global mean lower-troposphere anomalies now show a further sharp drop, lengthening the New Pause by three months, from 5 years 7 months last month to 5 years 10 months this month.
The HadCRUT4 data show no warming in the 6 years 9 months May 2014 to January 2021:
It is likely that the temperature anomalies will remain below the trend-line for another month or two, lengthening the New Pause still further. Forecasts for the rest of the year suggest that the present ENSO-neutral conditions will remain till the end of the year, with the possibility of another la Niña at the end of this year. The likelihood of El Niño conditions is thought to be remote for now.
It’s still early yet, but the Pause Deniers might want to dust off all their dog-ate-my-homework excuses for it. I’m not sure if they reached 100, but I think they came close. Who knows, maybe they’ll think of some new ones this time. Aliens?
Nick Stokes has shown this more compellingly elsewhere in the thread, but to reiterate the same comment I make every time Mr. Monckton makes one of his new “pause” updates, the trend for the past 70 months is not statistically significant:
Temperature data are very noisy, and over short timescales that noise can swamp the long term trend underlying it. The longer the time span you look at, the more confidence you can place in the observed trend. When you look at the longer term trend in the UAH data, it is undeniably warming.
Strawman much? No one claims it is statistically significant, nor is any claim made that it affects the long-term trend, that we have warmed a bit since the LIA (big deal). Pauses, especially the longer they get are of interest because they add a further layer of doubt onto the CAGW ideology, and the longer they get, the more frantic the Alarmists get, and the more wild their “excuses” get, which is a great amusement for Skeptics/Climate Realists.
You can achieve a flat trend both in data that have no trend (are composed of pure random noise) or data that do have a trend but have a low signal/noise ratio over the time period considered. Here I made a series with a 1:1 trend + some random noise. Over 100 “years” there is a clear and significant trend:
But if I pull out the first ten “years” there is an apparent flat trend that is not statistically significant:
If you were to bet that that 10 year “pause” was going to continue, or that it signaled a slowdown or end to the “warming,” you’d be out money.
There is no “layer of doubt” added by “pauses” like this one, since in any data that are comprised of signal+noise you can find any trend you want if you pick the right time interval. You can’t necessarily find a trend that provides any useful information about the behavior of the series by doing this, however.
You just don’t get it, I guess.
I certainly don’t get why Mr. Monckton is making claims that the data don’t support, no.
Weekly_rise should ask his paymasters for a payrise. I have not made any “claims that the data don’t support”: I have stated that the least-squares linear-regression trend on the UAH data shows no trend in the past 5 years 10 months. Of course, calculating least-square trends is beyond Weekly_Rise’s competence, or he would not have suggested I was making unsupported claims.
I suspect that the paid trolls here are whingeing even more than usual because they know how very powerful the last Pause proved to be as a way of shutting down climate fanaticism, and they are terrified that the same may happen again as the present Pause lengthens.
However, in this series I make no predictions: I merely report the fact – and it is a fact, however uncongenial that fact may be to Weekly_rise’s paymasters – that there is a Pause.
The claim you’ve made that is unsupported by the data is the claim of a pause. You simply cannot establish that based on the data you’ve shown. You need to calculate the uncertainty interval of the calculated trend for the data and determine whether a trend >0 falls inside of it (and I’ll give everyone in the thread a little spoiler: it is going to).
As I showed, you can find 10 year + “pauses” in artificial data with a baked-in trend.
MR. rise: You say the claim is “unsupported” by data; but you also say the problem is short time scales. In other words, the data DO support his claim, but you say the time scales are too short. Try to be precise: Do the data support it but you disagree with his method? Because you have shown it is supported by data.
You (and Bellman and Stokes) are simply being obtuse, using statistical sleight of hand (we know you can shift dates to show trend lines, but you can’t shift “today”) to obscure the all-too-obvious point: If CO2 goes up and the temp does not, then AGW is falsified.
Won’t disagree with your last sentence, I’m guessing you can even find a 10 year pause in Gavin Schmidt’s artificial data with a baked-in warming trend.
The data do not support the claim Mr. Monckton is making. The margin of error in the trend for the past 6 years is too large to claim that there is a pause in the warming trend. Is this clearer for you?
This is untrue, since CO2 is not the only thing influencing weather or climate. We expect a long term increase of the surface temperature in response to rising CO2 concentration, we do not expect there to be no year over year or month over month variability.
Your own work shows that there is no statistical support for the hypothesis that there is a trend. That is, the null hypothesis is to be accepted, which is that there is no relationship between the independent variable (time) and the dependent variable (temp). That is, the slope is zero!
Failing to reject the null hypothesis does not mean that you have proven it.
Let’s consider the semantics. If I give you a gift, and you don’t reject it, can’t I assume that you have accepted it, even if you hate it?
The alternate hypothesis is that there is a trend. That is, the absolute value of the slope is greater than zero. If the null hypothesis is rejected, then one generally accepts the alternate hypothesis. What are the choices? That there is either a trend or their isn’t! If the null hypothesis cannot be rejected, then one should accept it as being the best explanation, particularly when the graph shows it is approximately zero, and it cannot be rejected statistically.
Your position is basically rejecting the idea that anything can be proven with statistics. As you would have it, 1) rejecting a null hypothesis doesn’t prove the alternate hypothesis, one only assumes that it is the best of two competing hypotheses; 2) not rejecting the null hypothesis leaves one with two competing hypotheses with no way to prove either one. Why would you or anyone use statistics to try to prove something?
“Absence of evidence is not evidence of absence.” I’ve already showed you that it’s easy to find periods with no trend in a series made up of random noise plus a baked in trend (I.e. we know the trend is there, because we put it there, we just can’t see it for the noise).
Weekly-rise has just demonstrated that Pat Frank is right: propagation of error in models’ outputs means that all of their predictions of warming fall within the statistical uncertainty envelope and are, therefore, meaningless.
You’ve missed the point again: why are these temperatures not rising for the past 5 years as atmospheric carbon dioxide has been increasing the entire time.
And more than ironically, what your statistics are saying is that there is no correlation of temperature versus time, which is exactly the point CMoB is making.
CO2 is not the only factor influencing global temperature, especially on short timescales. There are other influences like internal variability, and natural and anthropogenic forcing that have an impact as well. This is my point, which you’ve missed. Climate data are noisy, and it is hard to make confident predictions about long term trends over very short timescales.
So on the basis of all this noise and uncertainty, we have to spend tens of trillions of dollars to stop using fossil fuel energy?
Be real…
The longer the period of observation, the lower the uncertainty in the trend. No one has suggested making policy decisions based on 6 years of data.
Mr. Carlo: I asked him to be precise, you asked him to be real. Maybe we are asking too much? ‘Cause I don’t think he will be either one.
I think you are correct here.
Mr. rise: Please do tell what “anthropogenic forcing” impact has been observed that “influences global temperature”? Remember my prior admonition- be precise.
I believe it is you who are missing my point: we don’t know if the temperatures aren’t rising because global warming has stopped or because the data are just too noisy to tell. The uncertainty in the trend over this small time period is too great.
Elementary considerations would lead us to conclude that global warming has not stopped. However, repeated long periods without warming are one way of pointing towards the underlying truth that the entire period of record since December 1978 shows global warming at a rate equivalent to less than 1.4 K/century, of which only 70%, or about 1 K/century equivalent (Wu et al. 2019, Scafetta 2021), was anthropogenic, and not the 3.4 K/century equivalent that IPCC (1990) had confidently predicted.
Nick Stokes has already shown us earlier that the long term trend shown in the UAH V6 data is about 1.4 deg C/Century, so there is no need to point toward this truth:
I am unsure why you’re suggesting that the IPCC “predicted” historic trends.
Also interesting to note, from the above graph, that the UAH series is, with the update of HadCRUT to V5 (the graph shows V4), quite an outlier.
Nick Stokes had no need to point out that the long-run trend on the UAH monthly data is close to 1.4 K/century equivalent: that fact had been made quite plain in last month’s column.
And it seems that the increasingly desperate “weekly-rise” has some difficulty with understanding the elementary physical concept of the arrow of time.
IPCC made its first medium-term prediction of anthropogenic global warming in 1990. That prediction was to the effect that in the following decades there would be anthropogenic warming equivalent to 0.34 K/decade, or 3.4 K/century. That turns out to have been an egregious overstatement compared with observed reality.
One understands that the continuing slow rate of warming, so very far short of the scandalously overblown predictions on the basis of which the climate fanatics got the current scare going, is vexing to them. But there it is: observation trumps prediction, so weekly-rise is just going to have to get used to it.
For the predictions were based on the false assumption that about 8 K global warming caused by preindustrial noncondensing greenhouse gases drove 24 K feedback response. The obvious error in that calculation is that it ignores the sunshine. Emission temperature drove nearly all of the 24 K preindustrial feedback response. Without much error, one can ignore feedback response in deriving equilibrium sensitivities today, so that there will be around 1.1 K equilibrium sensitivity to doubled CO2 concentration – and that is hardly a problem, now, is it?
You are conflating two different questions: 1) Is there statistical evidence for an increase in the trend over the last 6 years; 2) If not, why not?
Question 1 stands on its own, without a need for an answer to question 2.
Because question 1 null hypothesis cannot be rejected based on statistical probability, the conclusion is that there is no evidence supporting a trend for what appears visually to be trendless.
Yes, noisy data might be responsible for this situation, as for any similar situation. However, one is left with the possibility that there actually is not even an embedded trend, and the speculated noise hiding the data is not valid. How does one prove which of the two hypotheses apply to question 2?
Here’s a different view of the same concept I’m trying to relate. Similar to Nick Stokes’ graphic above, I’ve calculated the trend starting in each year through 2020. I’ve also plotted the 95% uncertainty interval for the trend in each starting year:
As you can see, when we get to very short intervals, the uncertainty in the calculated trend skyrockets. But note that the uncertainty in the trend doesn’t include 0 (grey line) until 2010, and at no point does the uncertainty in the trend exclude the trend since 1979 (red dashed line).
Mr. Monckton can keep updating his “pause” as long as he likes, that “pause” won’t be meaningful until he can show that the uncertainty in the trend excludes a positive rate of warming.This is why we can’t say that no evidence of a trend is equivalent of the presence of evidence for a non-trend over this period.
The satellite record was negative (-0.01 C) for March.
I am pretty concerned that those professionals, who should follow the global temperature trends and the reasons behind the changes, are not on the map at all. La Nina is too weak to cause temperature drops started last December. The reason is the weakening of shortwave tradition anomaly. This anomaly caused 50 % of the user El Nino 2015-16 and has been at the level of 1,7 W/m2 till 2019. It is the magnitude as CO2 forcing from 1750 to 2011 according to IPCC:
Here are the simple facts based on the CERES observations. The pause of the 2000s is back! The temperature of March is exactly the same as the average temperature of 2002-2014..
https://www.climatexam.com/blog
Hold on. I thought the old pause started in 1997 and lasted till 2017. But the new pause started in 2014?
I’m confused.
There is no exact rule. You get the same result if you start from 1997 or 2002.
the concept of pause is not realistic : nature doesn’t know it, it’s changing all the time depending on many different cycles from a few years to millions ! But during human life time, one could consider to live in a stable climate , which is NEVER the case !
To those unfamiliar with elementary statistics, the derivation of a trend on a data series must seem meaningless. However, the trend is correctly calculated. The climate is indeed inherently thermostatic, which is why there has been little more than 3 degrees’ variance either side of the 800,000-year mean (Jouzel et al. 2007). Over the past six or seven years, there has been a zero trend in global warming. So short a trendless period is not in itself statistically significant. However, if there is no la Nina for two or three years the present pause could become significant, which is why we track it here.
Hide The Paws