How the world really warmed between the 70s and the 00s, Part I

It bores me to death ceaselessly having to argue against assertive warmist claims about effects, simply presupposed as real, but – unfailingly – never supported by observational evidence from the real world, of altogether hypothetical mechanisms whereby increasing amounts of CO2 in the atmosphere is said to somehow warm the surface of the earth by radiative means.

It is the perfect circular argument. The perfect corruption of the scientific method. They don’t have to find and show at all that their claimed mechanism is working as postulated, because they already know it does. In advance. It’s there. Behind ‘the natural noise’.

‘Discussing’ this topic with the warmists, on their preset terms, from their compulsively linear (that is, CO2-bound) world perspective, thus gives as much meaning as arguing about the biological link between unicorns and horses.

This is why I thought I’d rather start myself describing and explaining how the global surface (and, by extension, the troposphere) really warmed since about 1970, i.e. ‘the modern global warming’, that period which is proclaimed with near absolute certainty by the IPCC to be mainly caused by our CO2 emissions. This is the period where allegedly the anthropogenic signal finally, clearly and undeniably stands out and starts overriding any natural drivers.

And I wanted to base my decriptive explanation solely on what the real-world data out there might reveal. On what Mother Nature herself might have to say on the matter.

This raw, unfettered method of actually following the data to see where it leads is seemingly abhorred by the warmists. All the ‘data’ that matters in their world comes as output from their models, fed and nurtured by the conjectures of their armchair hypotheses, by their predetermined answers … in short, their opinions about how earth’s climate should work.

This analysis, however, will be firmly rooted throughout in the observational data, in the natural processes and mechanisms that we actually know have an effect on the climate, regionally and globally … simply because we can see and track them directly at work in the real-world data.

In other words: We let reality itself tell its own story rather than our preconceived ideas about it.

The conclusion about what drives what comes last, not first.

That is, we adhere to ‘The Scientific Method.’

Unlike the warmist approach …

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

It all started out with my having a discussion with a warmist believing in the gradual, near-linear increase in ‘radiative forcing’ from rising atmospheric CO2 to be a (the) driver of global temperatures since the 50s. I asked him ‘Where specifically in the temperature record do you see this alleged ‘forcing’ operating?’ Could he somehow point out the CO2 ‘forcing signal’ to me?

He seemingly didn’t even understand the question. Point out the signal? It’s right there in front of you. You see? Temperatures have gone up. And so has CO2. There’s your ‘signal’.

In order to grasp such a reply, you need to understand the warmist mindset. They just know their hypothesis is right. Even before it’s been empirically found to be so. In fact, they’re so sure that, even if it could never possibly be empirically verified, in their minds it would be correct all the same. Purely on its perceived theoretical merits. No matter how much real-world data comes in to seriously question its core premises, it would still be correct. The data simply must be wrong somehow. Or needs to be ‘seen in a larger (or the right) context’. ‘Correctly interpreted.’ Because they just know that their postulated warming mechanism must work. Because … it has to. It can’t not. And as a result of this deep-rooted, unwavering faith in the veracity of a hypothesis whose pretty extraordinary claims are completely and utterly unsupported by any real-world evidence whatsoever, they feel (I would almost say, morally) entitled to cut short by leapfrogging the all-essential testing part of the scientific method and simply sit down and take any kind of warming for granted as proof they’re right. ‘Global warming’ equals ‘anthropogenic global warming’. Period. They have an absolute obsession with linear trend lines. They mentally project them onto every temperature graph they see. Because they just know in their hearts that such a trend line (in the long term, mind you) represents not only the gradually increasing ‘radiative forcing’ from rising atmospheric CO2, but also the effect it has on global temperatures.

You see the pigheaded circular reasoning at work here …?

A ‘trend’ in the warmist bubble world is itself a mechanism for warming, a physical process in its own right. A driver. Even though we all know that what a trend really is, is simply ‘the general direction of a collection of data’, oftentimes depicted by an applied (statistically generated) ‘trend line’. The data comes first. The data just ‘is’. The ‘trend’ is then simply one of many statistical tools that we humans use to describe it, to try and make some sense of it, a sense of order, a sense of pattern. That’s what our species does. What we need to do. It’s our instinct. The trend is not itself a part of the data. It’s one of the things that we read from it.

The warmists have taken this very human instinct to the extreme. They habitually discard the underlying data. They don’t see it. Don’t want to see it. Only the trend means something. Only the trend gives them answers. The warmists simply can’t be bothered with the data as it is. It’s only in the way. Noise. What they see is their trend line plastered on top of it. The background data doesn’t matter. The perceived trend is what counts. As long as it’s going up, that is … As soon as that’s no longer the case, you can be sure of one thing: They begin scrutinising the data. The very data they summarily ignored the moment before. Trend is no longer important. There are reasons for why the data progresses like it does, for why it doesn’t continue to go up, you see. You need to look at what’s behind the trend. See what ‘other potential drivers’ it reveals. The data. You’re a fool for not seeing the complexity … Adorable, isn’t it?

Well, I tried to get my particular warmist to be a little more specific. Yes, it’s a fair bit warmer now globally than it was 50 years ago, agreed. And that’s intriguing. But how and where – in the real-world data before you – do you see that the warming is caused by rising atmospheric CO2 levels and not by something else?

I then presented him with the following four segments of the global temperature curve (HadCRUt3), covering the last 57 or so years between 1957 and 2014, running next to the Mauna Loa CO2 concentration curve from 1958:





I then asked him again: ‘Where in these four plots do you see your mechanism operating? +CO2 -> +T. Anywhere? If it is there (and you seem all too certain it is), you should be able to somehow point it out, shouldn’t you?’

As you can all plainly see – it’s right there in front of you, after all – there is hardly a hint of an upward trend in any of the decadal temperature segments here (the first one clearly trends down, the other three for all intents and purposes have no trend at all), while the CO2 curve rises steadily and firmly in all of them.

So where is that conclusive CO2 warming signal? Where precisely can it be spotted? The dead +CO2 -> +T giveaway. The atmospheric CO2 ‘forcing’ mechanism at work on global temperatures, as a direct causative agent. The gradual rise behind or on top of the natural fluctuations? Why is nothing happening across time segments spanning 10, 15, 20 years? Why is there no response? At all? Not even a small one?

He objected, of course, to my ‘deceptive’ splitting up the long term curve.

The truth is after all, there has indeed been significant global warming during the period in question (1957-2014). No one denies it. But none whatsoever if you simply were to put the four split up segments of it (above) right next to each other.

So what’s going on? The answer of course gives itself: The total temperature increase must all occur … between the segments:


This is the first major data observation hinting at a different driver than the gradual increase in ‘forcing’ from rising CO2.

The entire rise in the mean global temperature since 1970 (and since 1957) is contained within three (3) steps or sudden upward shifts. And only there.

This observational fact alone should give anyone pause for thought. How come this is the way global temperatures consistently progress through time – flat decadal plateaus separated by sharp rises (or falls, like in 1964) inside the span of a year?

Global warming since the 50s did not happen along a steadily rising background trend. The trend comes after. Applied artificially. Methodologically. Mentally. The perceived ‘trend’ wouldn’t be there at all if it weren’t for the three sudden steps. ‘Modern global warming’ occurred at three distinct instances: in 1979, in 1988 and in 1998. And at no other time!

Why these particular years? I’ll explain. Or rather, I’ll let the data explain it for me.

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

The time series segments shown above do not run exactly between the three actual temperature shifts. The precise timing of these shifts are in fact quite easy to pinpoint. And I will show you how.

The first (and most decisive) of the three global warming steps occurred in 1979, on the heels of what has been called ‘The Great Pacific Climate Shift.’

The second step occurred in 1988 and the third (and thus far final) step took place in 1998. Both of these sudden hikes are directly associated, like the first one, with major climate regime shifts in the Pacific basin. However, the latter two are of a different kind than the first one. This is something that will be explained.

How, then, can you tell that these step changes in fact occurred when I say they occurred and only then?

It is all perfectly straightforward.

In the following, I’ll present you with some graphs. I want you to study them closely. They portray the monthly anomaly evolution of two different temperature datasets going from 1970 to 2014.

The first dataset is ‘HadCRUt3 gl’ which is the land+sea weighted average, that is, the global instrumental surface record. There are obvious reasons why I’ve chosen HadCRUt3 to represent this metric over GISTEMP, NCDC or HadCRUt4. They all involve the practice of consistent and systematic (targeted) upward data adjustments in the latter three, most of them after 1998, all of them post 1990. To me it is pretty much self-evident why HadCRUt3 is the most believable rendition of actual global temperature evolution since 1970. But I will not get into demonstrating it here (it wouldn’t be much of a problem, though a fairly extensive exercise …). That’s for another time, another post.

(There is actually a major specific error contained also within the HadCRUt3 dataset. But this is easily spotted (though it has strangely (?) never been addressed, much less amended, by HadCRU themselves, most tellingly not even when they introduced their new and ‘improved’ version 4, to make their own curve look more like (almost identical to) NASA’s – with this, they rather seemed to go to great lengths to try and cover up the error) and thus easily corrected for, the significant artificial jump in mean anomaly level across the inadequately calibrated seam between two different data sources for the Hadley Centre’s SST product (HadSST2), occurring at the 1997/98 transition.)

The second dataset is ‘Kaplan+OI.v2 NINO3.4 SSTa’ which represents the East Pacific portion of the ENSO phenomenon. Kaplan is used up to and including Oct’81, from which point Reynolds OI.v2 takes over. A decent overlap has been applied in generating this composite dataset, to make sure the fit is near-perfect (the two separate datasets are impressively (and reassuringly) compatible to begin with).

The latter dataset is then downscaled significantly (divided by 6.84*) to match up properly when superimposed on the former dataset. Note that none of the time series are lagged relative to the other. They are compared directly and synchronously.

*The NINO3.4 anomaly data is divided by 8 to compare satisfactorily with global SSTa data, only by 4 when compared with land data. Since there is 71% ocean and only 29% land on earth, the mean ratio when comparing NINO3.4 with global land+sea becomes 6.84.

I have split the total series (1970-2014) into four overlapping segments:

  • Jan’70 – Feb’88
  • Jan’79 – Mar’98
  • Jan’88 – Apr/May’14
  • Jan’98 – Apr/May’14

This I have done for a very specific reason. Each of them (except the last one) spans across a global step in temperature anomalies relative to NINO3.4 and depicts the periods before and after the steps. Again, there are only three global steps in total: one in 1979, one in 1988 and one in 1998:


Blue curve: HadCRUt3gl (adjusted down post ’98 by 0.064 degrees, see about the need for this correction above); red curve: Kaplan+OI.v2 NINO3.4 (divided by 6.84).

The first segment, then, stretches from 1970 across the 1979 global step and all the way to just before the 1988 step is about to be taken. The second segment stretches from the 1979 step, across the 1988 step and down to just before the 1998 step is about to occur. The third segment stretches from the 1988 step, across the 1998 step and down till today (2014). The fourth segment stretches from the 1998 step and down till today (2014), with something that could look like a so far failed ‘attempt’ at another step up.

Each of the three verified global steps occurred within the short periods covered by the three vertical green columns (above). You simply can’t miss any of them. They are all so evident and their timing so specific that it’s almost worth a chuckle when you see it!

There is distinct and significant global warming (that is, warming outside the NINO3.4 region in the equatorial (tropical) zone of the East Central Pacific Ocean) at each of the three steps. At other times there is no hint whatsoever of any decadal global warming above NINO3.4 (that is, no gradually increasing upward divergence before, after or between the steps; none!).

Segment 1 (Jan’70 – Feb’88):

Step 1

Follow the global and the NINO3.4 curves from Jan’70 to Jun’79 (dark green line in the middle of the green column, global step 1). Look how closely they track each other, only with the latter leading the way, conspicuously at the big events, not so much at other times. There is absolutely no global increase relative to NINO3.4 from 1970 to 1979. No hint. No trace. Then observe what happens across the dark green line. Global temperature anomalies place themselves comfortably above NINO3.4, but without any further divergence after the step. If we move on to Segment 2, we will see just how closely they once again follow NINO3.4. Everything happens abruptly at the 1979 step alone. Remarkable, isn’t it?

Segment 2 (Jan’79 – Mar’98):

Step 2

We have adjusted the NINO3.4 curve up, to once again lie directly superimposed on the global curve. Now follow their course together from Jul’79 (first dark green line) to Feb’88 (second dark green line, inside the green column, global step 2). Once again they track each other closely and neatly. You will notice the El Chichón global volcanic impact in 1982-83 just before the El Niño of 1982/83 struck, and also the fairly weak global response to the El Niño of 1986/87. With the directly following 1987/88 El Niño, though, global anomalies catch up, only with a small delay. The global lag at the big ENSO events appears to be smaller after the 1979 shift than before it. Something big occurred in the late 70s. What? ‘The Great Pacific Climate Shift’ of course. With it, the global response to ENSO changed fundamentally. But that’s another story for another day.

What happens in the wake of El Niño 1987/88? Past the dark green line? The NINO3.4 SSTa drops like a stone, into the mighty La Niña of 1988/89, one of the strongest on record. Global temperatures, however, fail to adequately follow, as they normally would. In fact, far from it. And by that, a new and higher global temperature level is created. However, once again there is no further global upward divergence relative to NINO3.4 to be observed after the specific step up in 1988 (refer to Segment 3).

Segment 3 (Jan’88 – May’14):

Step 3

Once more NINO3.4 is adjusted up to fit on top of the global curve. And the same thing happens again. Global and NINO3.4 temp anomalies follow each other tightly all the way down to Mar’98, just after the peak of El Niño 1997/98 (second dark green line, inside the green column, global step 3). The global volcanic impact of Pinatubo is clearly visible between late 1991 and ’94, but other than that, the two curves do not diverge to any meaningful degree, NINO3.4 just leading by a small amount.

But what happens in the wake of El Niño 1997/98? Past the dark green line? Exactly the same as what happened in 1988. NINO3.4 plunges into La Niña 1998/99, but global temperatures do not adequately follow. From this point on, starting with the year 1999, they stay well above NINO3.4, but show no hint of a further increase down till today.

Which leads us to the final segment, number 4 (Jan’98 – May’14):

Post Step 3

Remarkable fit once again. A small lead by NINO3.4, but no general global increase relative to NINO to be found. And certainly no gradual global upward divergence from 1998/99 to 2013/14.

However, something did happen during the waning of El Niño 2009/10 into the deep La Niña directly succeeding it that so closely resembled what happened after the similar El Niños of 87/88 and 97/98, that one could easily have come to expect the establishment of yet another raised global temperature level in its wake. That level can so far not be said to have materialised. But one thing’s for certain: We live in interesting times …

(What happened in 2010 that so closely resembled what went on in 1988 and 1998, you ask? I’ll come back to that later.)

So what does this exercise tell us? It can tell us something about processes. How the earth system works. What really caused ‘global warming’ from 1970 to 2014? The answer is as simple as it’s fundamental to all science: Let the data lead the way. Let it speak.

Whatever caused the three specific and abrupt global steps up relative to NINO3.4 – in 1979, 1988 and 1998 – caused ‘global warming’ from 1970 to 2014.

Now, could a gradual increase in a theoretically hypothesized ‘radiative forcing’ from a mounting global concentration of CO2 in the atmosphere have effectuated these three sudden instances of significantly elevated mean global temperatures? Hmm. One could by all means still put this forward as an argument, but one would surely have a hard time substantiating it, backing it up with relevant empirical evidence from the real earth system.

You see, global temperatures simply show no sign whatsoever of any increase relative to NINO3.4 from 1970 all the way down till today except during just these three sudden, distinct upward shifts. And the tight (dependent/causal) connection to NINO3.4 is obvious the entire time … outside the steps.

So what exactly happened in 1979, 1988 and 1998? That’s what we need to find out. That’s what needs to be explained.

And I can assure you, it’s all process-based, all caused by internal natural processes. And it all originates in the Pacific Ocean. But that’s for later …

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Take a look at these graphs. I have divided the total time series into the precise segments between the global steps, equivalent, but not equal to, my original segments from upthread. There are no time gaps between the segments. Red curve is still NINO3.4 SSTa, blue curve still HadCRUt3gl. As you can see, I’ve generated the corresponding linear trend line of each dataset. Note how the NINO trend is always steeper than the global within these particular segments:


So what’s my point?

It is easily shown above that the tilt in the linear trend lines drawn across the data is completely dependent on what the data actually does between the chosen endpoints.

A linear trend (expressed by its trend line) is, after all, once again nothing but a statistical tool with no real interpretive skill, only descriptive. What do I mean? Let me give you an example. If we draw the trend lines across one (or two, or all) of the steps (’79, ’88 or ’98), then the global trend will always be steeper than the NINO3.4 trend. If, however, we draw them between the steps, like we’ve done here, then the NINO3.4 trend will always be steeper than the global. In this lies the recognition that statistically implemented linear trend lines tell us nothing about the physical processes at work, what caused the trend to be what it is. If you throw away the data behind the trend and just keep the trend line, then you’ve thrown away the means by which you would be able to actually find out why (and how) the temperatures rose. That is, if you do not already believe a priori that the answer is given by the (rising) trend line itself …

Science vs. ‘climate science’.

Half the time trend lines can’t even tell us whether one dataset has truly risen above another from start to finish. Look at the superimposed curves above and the trend lines accompanying them. What do they tell us? That NINO3.4 is rising above and away from the global curve? Strictly according to the trend lines it does. But if you actually look at the data behind, the plots themselves, you see why the former has a steeper trend than the latter. Take the segment 1988-98 for instance. It starts with a La Niña and ends with an El Niño. The global curve normally exhibits a muted ENSO signal as compared to NINO3.4 at the big events, so the latter will go relatively deeper with a cold event and higher with a warm event. On top of this, we get the abnormal global warming (outside the NINO region) during the ’88 step, meaning, also for this reason the global curve traces significantly above the NINO curve for the segment’s first year. This is what produces a global trend gentler than the NINO trend. There is no gradually increasing divergence. NINO isn’t really warming faster than the global. You can easily observe how the two curves follow each other just as tightly in the second half as in the first half of the graph.

Exactly the same goes for the segment of 1998-2010.

Likewise, we know – or at least should know – by now that the only reason why the global trend is steeper than the NINO3.4 trend between for instance 1970 and 1988 is that there happens to be a hurried yet significant global step up occurring right in the middle of the period in question (1979). Both before and after this step, after all, NINO3.4 carries a steeper trend than the global.

So there is no discernible ‘background trend’ anywhere. The warmists turn everything on its head by claiming there is. It goes directly against the data. There is only the three global steps. They are the sole reason that the global rises above NINO. The entire ‘modern global warming’ is contained within them. There simply is no escaping it.

This fact alone should constitute a very real problem for the ‘increasing CO2 forcing causes global warming’ position.

But it gets worse. For we have yet to actually look into the three steps. To explain them. Where and how did they originate? Any CO2 fingerprints coming up? Not at all. Something else entirely? Definitely.

Again we go back to the data …

7 comments on “How the world really warmed between the 70s and the 00s, Part I

  1. Ingrid says:

    Gratulerer med ny blogg – gleder meg til å følge med.

  2. Pretty nice post. I just stumbled upon your blog and wished to say that I’ve truly
    enjoyed surfing around your blog posts. After all I’ll be subscribing to your feed and I hope you write again soon!

  3. Jess says:

    Gleder meg også! Reading through your explanations my old calculus classwork started coming back. Once again, I see that many graduates fail here to assimilate the concepts behind the mathematical representations. Same thing I found in Electronic Engineering. Too many engineers skilled in the mechanics (computer sims, graphing, models, etc.), but don’t understand how to apply them. And way too many who could not even remember the basic theorems! The Keystone cops have all been ceremoniously given hammers, and let loose to build houses without an experienced Foreman.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s