Guest post by David Archibald,
The solar plasma temperature has plunged to a new low for the instrument record. Coincidentally or not, the temperature of the southern hemisphere has also plunged over the last couple of weeks. When do we start worrying?
Figure 1: Temperature of the solar wind plasma
As Figure 1 shows, the temperature of the solar wind has hit a new low for the instrument record. As it is energy from the Sun that keeps the Earth from looking like Pluto, the lower plasma temperature indicates that the Sun’s surface is cooling. Surely the Earth’s surface will follow.
Figure 2: Alpha particle to proton ratio in the solar wind
Similarly the alpha particle to proton ratio has hit a new low for the almost 50 years of the instrument record. The decline for the peak ratio in each solar cycle is even more dramatic. The question that naturally arises is this: Is there a lower bound for this ratio?
Figure 3: Solar wind flow pressure
Solar wind flow pressure has hit a new low for the instrument record. There a couple of interesting things about this chart. Note that the lows for the last three solar cycles are aligned as indicated by the blue line. This implies that there is a disciplined process involved. Note also the low activity in the late 1960s that set up the 1970s cooling period.
It is the solar wind flow pressure combined with the Sun’s magnetic field that reduces the flux of galactic cosmic rays reaching the Earth. As these two parameters we can expect a spike in the neutron flux about a year from now. In turn that is expected to increase cloud cover and the Earth’s albedo.
Figure 4: Kp Index
The Kp–index is a global auroral activity indicator on a scale from 0 to 9. What is evident in this graph is the change in regime from the Modern Warm Period that ended in 2006 and the New Cold Period.
Figure 5: Ap Index 1967 to 2020
According to Omniweb’s data the Ap Index has also hit a new low for its data record.
Figure 6: 2 metre temperature anomaly for 2020
Figure 6 is from Oxford academic Karsten Haustein’s website. It is updated daily. It shows that the temperature of the southern hemisphere (the blue line) has plunged 0.6°C in the last couple of weeks and is continuing to plunge. Could it be that the new lows for some solar parameters is having an instant response? The Antarctic plateau is the Earth’s refrigerator.
David Archibald is the author of The Anticancer Garden in Australia.
So, NOAA reckons this is how “average global temperature” is derived –
Other sources say that the methods used to calculate average temperatures are really “very simple”.
I would say that was a typo – what they meant to write was – very SIMPLISTIC
Now, at the risk of sounding like a contrarian just for the sake of contrarianism, I suggest that the “average global temperature” construct has so many flakey inputs, that the whole result is risible.
I don’t wonder at all that after some 40 years of multi-billion $ annual spending on climate “science”, employing tens of thousands of researchers, the closest ‘official’ prediction of future global average temp rise is an unconvincing construct of 1.5C – 4.5C, or thereabouts.
The “global average temperature” house has very unsound foundations. No wonder the outlook from the front porch is so blurry.
Also, if IPCC et al were listed trading stocks, and their reports were being offered as prospectuses for investor guidance, the corporate regulators would have had their whole boards and management people serving long sentences in jail by now for deceptive & misleading assertions.
You aren’t a contrarian.
If I tell you the average temperature is 35degF can you tell me what the maximum temperature in the data set was? What the minimum temperature in the data set was?
If you can’t tell what the maximum and minimum temperatures are then exactly what is the average telling you?
Are the maximums going up? Are the minimums going up? Is it a combination? How do you make realistic decisions if you don’t know that the temperature envelope is doing? Will you need more heating 20 years from now? More cooling?
There is nothing to worry about. It is settled science. The sun has nothing to do with the climate. Only CO2 controls the climate. The sun is irrelevant.
I thought it was getting cooler….
https://breadonthewater.co.za/2020/07/07/brrr-it-is-getting-colder/
I wouldn’t use the adjusted HADCRUT 4 data for any serious analysis of temperature change.
Ja. Ja.
https://breadonthewater.co.za/2020/07/07/brrr-it-is-getting-colder/
David Archibald, I guess I’m the only person here who cares enough to try to replicate your data … sorry, I couldn’t. Here’s your Fig. 1.
I went to the OmniWeb site and asked for the same graph, same subject, same averaging, and I got this:
Note that unlike your graph, the troughs never go much below 5E+4, and are not decreasing over the period of record. Not sure why the difference, I encourage people to try it for themselves. Looks like maybe there is no difference, you just put a bogus line underneath your graph …
Finally, if you look at daily averages rather than 27-day averages, you’ll see that many days the temperature goes down to almost zero … surely you don’t think that is somehow representative of the solar temperature?
w.
Here are the daily averages along with a 360 point gaussian average … not seeing any reason to think that this will somehow cool the world.
w.
Willis,
The data ranges over 3 orders of magnitude, so the plot will behave better if you plot log10 of data.
Also, at 1 AU, the SW proton temperature is not really a good estimator of solar temperature (i.e. when the temperature of the SW when it left the solar surface). SW expands non-adiabatically as it spirals outward, with significantly less cooling observed than would be expected from an ideal gas expansion. So some other processes are involved in this heating. So you cannot look at a dip in the graph and say “The Sun is cooling!”
Daniel Vech et al., “Nature of Stochastic Ion Heating in the Solar Wind: Testing the Dependence on
Plasma Beta and Turbulence Amplitude”
https://iopscience.iop.org/article/10.3847/2041-8213/aa9887/pdf [2017]
There is only one difference between your graph and Archibald’s, and that is the last point. Since it is a 27-day average, didn’t occur to you before saying this that the difference is due to the graph being called on a different day and that affecting the average that constitutes the last point?
Geez, what a blunder, Willis. Making an issue for one point out of more than 720. The downward trend is unmistakable.
Making an issue for one point out of more than 720.
Nonsense, it is Archibald who claims that the last point is a 29-day average. It is not. It is the average of only a couple of days into the 27-days. A typical Javier blunder.
Where does he claim such thing? Please exact quote.
You are just making things up as usual.
Archibald just got the graph from Omniweb and never noticed that the program returned an incomplete average for the last point. I don’t think 99.99% of the people could have noticed that. Willis on the contrary failed to find such an obvious explanation for why there was a disparity between Archibald’s graph and his.
Apart from my obvious typo, all Archibald’s graphs say 27-day averages.
I don’t think 99.99% of the people could have noticed that
It was obvious, so almost everybody [who would care to look] would notice that.
Willis deliberated omitted August not to fall into that trap. I would not call it a ‘failure’ to expose Archibald’s failure to do due diligence
OK, so Archibald’s claim turns out to be Omniwebs’ claim since it is something that Omniweb and not Archibald put on the graph.
Once more you show us that what you say is biased and not to be trusted.
It is obvious in hindsight, once the issue has been raised and investigated. I had no idea the last point in OMniweb’s averaged graphs is an artifact. It is illogical. They shouldn’t do that.
Willis fell in one trap after another, all of them self-inflicted. He first mistook rotations for months, and he then cherry picked an ending of his liking without telling anybody so that his last point was his particular artifact (a 17-day average) instead of Omniweb’s.
Willi’s graph from Omniweb also says 27-days average. Do you also believe he is making a wrong claim? Your bias shows.
Javier August 24, 2020 at 4:59 am
Once Archibald went and got the graph WITHOUT doing what I did, looking to see if the final data was a full month’s worth, and he published it, implicitly saying it was true, and he used the partial month as his endpoint for his bogus “trend line” … at that point, it’s absolutely David’s claim.
It may be illogical but it is common, common enough for me to look for it.
Huh? That makes no sense. Here is the tail of the OMNIweb data I used:
The last point I used was day 196 of 2020, which is July 15th. It contains 27 data points … where is this 13-day “artifact”.
That makes absolutely no sense.
Look, Javier, here’s the bottom line.
First David took a graph off of the web without looking closely at the data. So he got a result I didn’t get … he got a very low point at the end. You swallowed that without even looking at or understanding the underlying data. Me, I didn’t make that mistake, nor did Leif. That seems to drive you nuts, but it is a fact. (David also cherry-picked the starting point, but that’s another issue.)
Next, to try to convince us that solar wind temperature was going through the floor, David used a spurious “trend line” that goes from his (cherry picked) first data point to his (bogus) last data point. Even you don’t engage in that kind of nonsense.
Next, David omitted looking at the actual trajectory of the temperatures, which ROSE from the start of the data to the 1990s and dropped since then. If his claim that the drop in the SW wind temperature would make earth cooler were true (it’s not), then why would the dropping temperature post 1990 affect the Earth but not the rise up to 1990?
Now, it seems that you made the very foolish mistake of EVER believing what David says without actually doing the hard work of going to get the data and actually looking at it. And as a result, you are twisting and wriggling trying to get out from under, not my mistake, not Leif’s mistake, not David’s mistake, but YOUR mistake—you believed without checking, and in particular, you were so immensely foolish as to believe David Archibald without checking …
Unfortunately, rather than just admitting that both you and David made foolish mistakes, it’s driven you into an insane frenzy trying to pin all of the blame for both your and David’s mistakes on Omniweb, Leif, and me.
Not gonna work, bro’ … Leif and I were the ones that got it right and pointed out the problems in David’s claims, and Omniweb is just a data provider—can’t blame them when David drives the data bus off a cliff …
My strong suggestion would be to find another hill to die on. You are defending a piece of very, very poor work, and attempting to blame Leif and I for your and David’s errors. Never gonna happen, people are only going to point and laugh. Find something worth your effort, you are wasting your abilities defending David Archibald.
w.
that his last point was his particular artifact (a 17-day average) instead of Omniweb’s.
Willi’s graph from Omniweb also says 27-days average. Do you also believe he is making a wrong claim?
For the last time: the text on the graph is Omniweb’s and the graph shows [per Omniweb] 27-day averagesalso for the last point, not a 17-day average as you think. The one making wrong claims is [as usual] you.
You keep getting it all wrong Willis. I didn’t swallow anything. I didn’t believe anything David Archibald said. I left it very clear in my critical comment above. I am not defending him.
The trend in a graph of over 700 points does not depend on the last point, obviously, but you decided to make an issue of it. All that Archibald had done was to post the graph provided by Omniweb. What a crime.
The linear trend is downward for the entire series. The highest year average is for 1974. The highest decadal average is for the 1973-1982 decade.
David is correct that plasma temperature has decreased since mid-70s.
There is no problem with Archibald’s data as you wanted us to believe. The last point was produced by Omniweb and is a non issue that doesn’t affect his correct claim of a downward trend. The problem with Archibald lies on his ridiculous claim that the 40-years trend in solar plasma temperature has anything to do with what temperature has done in the Southern Hemisphere for the last two weeks. There is zero evidence for that and I don’t buy it.
The linear trend is downward for the entire series. The highest year average is for 1974. The highest decadal average is for the 1973-1982 decade.
David is correct that plasma temperature has decreased since mid-70s.
What is wrong with this is that the solar wind has many properties that are all correlated. The temperature varies the most due to shocks and flares. One has to take the variation of all the components into account [Archibald tries to do this by presenting many graphs] . Overall, activity was low in the mid-1966 and rose until the mid-1990s and has since decreased again to about the level it came from, so there has been no long-term trend in the solar wind. To believe otherwise and to attach any significance to such belief are marks of amateurs with no understanding of the science.
Here is a plot of yearly averages of proton temperatures [dark blue symbols] and their standard deviations:
Data before 1971 [marked by shading] are less reliable than later data. The standard deviation [red squares] are related to the temperature averages through the relationship sigmaT = 1.4 T with an R^2 as high as 0.7. A few outliers are marked by crosses. If you scale the standard deviation up to T [using 1.4*T] you get the open pink squares so you cn se how good much to trust the scaling. The bar in the lower part of the Figure shows that there is no long-term trend, as we actually would expect because other solar indices also don’t show any such trend.
No, you didn’t understand the problem. You also said:
so that his last point was his particular artifact (a 17-day average) instead of Omniweb’s.
Not the case.
This data point does not contain a 27-day average. It contains only two points in the average. That is what the 2 in the fourth column means. This is the data point that made the difference between Archibald and Willis graphs. So I rightly identified the problem.
No, you did not identify he problem. You did not know about the 2 days or the fourth column. You thought it had to do with ending at the end of July. So, you were sloppy, and later tried to take credit for something you did not identify correctly. And instead used it to beat on Willis [‘blunder’ etc.]
The data:
Selected parameters:
1 # of points in Plasma averag.
2 SW Plasma Temperature, K
YEAR DOY HR 1 2
2020 7 0 27 55501.
2020 34 0 27 67195.
2020 61 0 27 58825.
2020 88 0 27 65422.
2020 115 0 27 39312.
2020 142 0 27 44134.
2020 169 0 27 39900.
2020 196 0 27 66177.
2020 223 0 2 25021.
Javier, please tell us how one single point justifies the trend line in Figure 1. You can’t, because it is BOGUS. Real trend lines don’t jump from start to finish and hit nothing in between.
Plus, learn to read. What I said was:
“Looks like maybe there is no difference, you just put a bogus line underneath your graph …”
And now you come to tell us there’s no difference, just the last point? Care to know why I left off the last point? Because my graph goes to the end of July, where his falsely presents the first ten days (or less) of August as if it were a full month.
Next, he’s cherry picking … the data starts in 1964 and he’s started his in 1967, without saying one word about that.
Next, he’s all on about how the minimum in the data shows that the world is cooling … so why doesn’t he say anything about what the MAXIMUM of the data shows? It started cooling down from the max around 1990 … where’s the corresponding cooling?
w.
Willis, you always ask people to quote your exact words to avoid confusions, except when you want to create the confusion yourself. I quoted the exact words I was discussing. They refer to your inability to replicate Archibald’s graph. That you now bring a host of new issues to the discussion is an admission that I am correct on that point.
Well, then you made a new mistake. 27 days averages are for solar rotations, not for calendar months.
You are just making your rant against Archibald sloppier and funnier than his article.
Then you are guilty of cherry picking yourself as you truncated the graph at the end of July and you didn’t say one word about that in your original comment. No wonder you couldn’t reproduce Archibald’s data when you were cherry picking the end point.
This is getting funnier and funnier.
This is getting funnier and funnier.
You are not funny, but ridiculous. Willis did not ‘fail’ or show ‘inability’ in replicating Archibald’s graph, but rather exposed the lack of due diligence Archibald showed in not noticing that the last data was not a [full or significant] 27-day average and therefore introduced a false trend. You could gain some credibility by not defending Archibald’s sloppiness, but, no, you join him in infamy instead. Poor behavior, not becoming a gentleman. Shame on you.
You are the ridiculous one. Willis confronted Archibald’s sloppiness with his bigger sloppiness.
While Archibald just took the graph given to him by Omniweb without knowing that the last point was an artifact introduced by Omniweb, Willis truncated the data cherry picking its ending on the 31st of July, wrongly believing that 27-days averages referred to full calendar months instead of solar rotations. Since Willis choice is day of the year 213, and the rotation goes to day of the year 223, the last point in Willis’ graph is an artifact of only averaging 17 days.
The difference in sloppiness is that in the case of Archibald the artifact was introduced by Omniweb, while in the case of Willis it was introduced by himself. Having the last point wrong would be inconsequential except that Willis decided to make a point attacking Archibald because of it. We have to thank Willis for the laughs his decision has brought.
I am not defending Archibald’s work. I have written one of the strongest critics in the comments. But Willis attack on his credibility for not being able to reproduce his data is entirely unjustified, as several commenters have noticed.
But I see the dynamic duo gets in action even to defend their own sloppiness. Shame on you.
the last point was an artifact introduced by Omniweb,
Not at all. Omniweb plots the data it has. It is up to the users to check if the data points are representative. Archibald and you didn’t. End of story.
Don’t get me involved. I have nothing to do with either Archibald or Willis graphs endpoints. I was the one identifying the cause for the disparity in the first place (August 23, 2020 at 2:51 pm).
The end of the story is that Willis made an unjustified attack on Archibald and in doing so committed multiple errors that he still refuses to acknowledge, while his friend Leif turns a blind eye to his errors and attacks anybody that points to them. Let’s recapitulate, Willis:
–Mistook months for rotations
–Cherry picked July 31st as the final day for his data request
–Obtained a final point that was a 17-days average instead of 27
And with that he went to criticize Archibald’s graph without any clue about the source of the disparity, that took me one minute to spot.
And through that he gets cheered by you and Steven Mosher that calls him the only real skeptic at WUWT. Tamino self-described as “Hansen bulldog,” perhaps you should self-describe as “Willis bulldog.”
This is the most hilarious thread I remember in WUWT, and the more you try to justify Willis the funnier it gets.
While Archibald just took the graph given to him by Omniweb without knowing that the last point was an artifact introduced by Omniweb, Willis truncated the data cherry picking its ending on the 31st of July, wrongly believing that 27-days averages referred to full calendar months instead of solar rotations.
You obviously do not know how Omniweb works. Even if you specify an end date of 31st of July, Omniweb gives you a plot [and a listing if you asks] including the last full rotation that includes the end date you have given, so Willis’ plot was perfectly correct.
Here are the data given by Omniweb:
Listing for omni2_27day data from 19631138 to 20200731
Selected parameters:
1 Bartels rotation number
2 # of points in Plasma averag.
3 SW Plasma Temperature, K
YEAR DOY HR 1 2 3
1963 360 0 1785 27 9999999.
1964 22 0 1786 19 9999999.
…
1965 142 0 1804 21 9999999.
1965 169 0 1805 18 9999999.
1965 196 0 1806 19 100853. <— this is the 1st rotation with data
1965 223 0 1807 22 94362.
1965 250 0 1808 24 72135.
…
2020 115 0 2547 27 39312.
2020 142 0 2548 27 44134.
2020 169 0 2549 27 39900.
2020 196 0 2550 27 66177. <— this is the last rotation with data ending 20200816
so no ‘blunder’ by Willis, but a BIG one by you.
I was the one identifying the cause for the disparity in the first place (August 23, 2020 at 2:51 pm).
No, your assertion was dead wrong. The disparity is not caused by a wrong end date, as Omniweb does not stop at the end date you give it, but goes to the end of the rotation regardless. Anther example of ignorance on your part.
I don’t have much use for Ommiweb data, but that doesn’t mean that Willis was correct as you say. His misunderstanding of solar rotations and his cherry picking of a final date remain as errors. His attack on Archibald for something that is due to Omniweb mode of functioning remains unjustified.
–Cherry picked July 31st as the final day for his data request
–Obtained a final point that was a 17-days average instead of 27
No, Omniweb rounds the plot up to the full 27 days if there is data there, regardless of the final day you give it.
See, you don’t know how this works and don’t seem to care enough to find out as long as you can attack someone.
I don’t have much use for Ommiweb data, but that doesn’t mean that Willis was correct as you say.
It means that his plot was correct.
His attack on Archibald for something that is due to Omniweb mode of functioning remains unjustified.
First, it was not an ‘attack’, but an attempt to understand what was going on.
You,r on the other hand, calling it a ‘blunder’ is a bona fide personal attack, which has come back to bite you, as Willis’ plot was correct and Archibald’s was not.
You simply do not have the scientific training or insight to pontificate on things you do not understand. You boasted that you understood and have applied the ‘scientific method’ in your work, but have failed to produce a link to your best work on the topic at hand, perhaps for good reason…
the last point in Willis’ graph is an artifact of only averaging 17 days.
this comment of yours shows that you don’t know what you are talking about.
Just repeating it is no good. Perhaps it is time to stop sounding like a broken record.
Which fact should have been obvious.
Hmm the only discrepancy I see is the end point.
The dip at the present is caused by a 27-day average only having a couple of data points in it:
Typical Archibald sloppiness.
True, Leif, as I found out when I went to replicate his claims.
w.
No. It is Omniweb sloppiness. If one asks for 27-day averages one should not get any average of less than 27 days as that introduces an artifact.
Archibald got not less confused by that than Willis, that cried he couldn’t replicate the data when Ulric and I had no problem spotting the issue.
one should not get any average of less than 27 days as that introduces an artifact.
Nonsense, as there are lots of 27-day rotations with less than 27 days of data:
when Ulric and I had no problem spotting the issue
and yet you claimed that ” I don’t think 99.99% of the people could have noticed that”.
Not before the issue had been raised and investigated. Afterwards it becomes easy to spot (not for Willis).
Not before the issue had been raised and investigate
By me, in fact. After I did that even you could see it.
You got that wrong also. I spotted the problem in my comment:
https://wattsupwiththat.com/2020/08/23/solar-plasma-temperature-is-plunging-should-we-worry/#comment-3067562
My comment is timed at 2:51 pm, yours at 3:45 pm, almost an hour later.
Claiming a precedence you haven’t. Shame on you.
I spotted the problem in my comment:
No you didn’t as what you ‘spotted’ was not the real problem, but was intended as an attack on Willis. Regardless of the end time, Omniweb rounds it up to include the whole rotation as I have shown.
Yes I did. I said the problem was that the last point was not a 27-days average as logic would dictate. The last value for plasma temperature given when one introduces the most recent available date includes only two points:
FORMAT OF THE SUBSETTED FILE
ITEMS FORMAT
1 YEAR I4
2 DOY I4
3 Hour I3
4 # of points in Plasma averag. I4
5 SW Plasma Temperature, K F9.0
2020 142 0 27 44134.
2020 169 0 27 39900.
2020 196 0 27 66177.
2020 223 0 2 25021.
That’s a curious behavior to say the list
It is clear you didn’t understand what Omniweb was returning
I understand extremely well what the OMNI data are and what the Omniweb returns.
I have worked with the OMNI data and the people curating the observations since the beginning of the dataset, as acknowledged by Joe King in “Interplanetary Medium Data Book Volume 77, Issue 4 of NSSDC/WDC-A-R & S” from the National Space Science Data Center (1977):”
Acknowledgements are also due to L . Svalgaard of Stanford University for his contribution
Knowledge of the data and its provenance is crucial in this business.
You, on the other hand, didn’t even care to look at the data, so are not qualified to even comment on them.
I said the problem was that the last point was not a 27-days average
No, you didn’t say that. You said:
“the difference is due to the graph being called on a different day”
and later believed [without checking!] that it was a 17-day average because you evidently believed the last day was the 31st July. Neither of these beliefs were true, so you did not understand what was going until I told you that the last day of Wilis’ last point was 16th August, because that is the [very logical] way Omniweb works when you ask for 27-averages.
Perhaps now you do after I explained it. But you didn’t when you talked about it with Ulric Lyons.
Then the more surprising that you would get this wrong:
You said that and it is wrong. Day 196 is 07/14/2020 The last full rotation ends on day 222, 08/09/2020, 7 days before you said.
As Omniweb shows the two days that enter into the last point are the 10th and 11th of August.
2020 222 0 52947.
2020 223 0 28892. <— last rotation starts (08/10/2020)
2020 224 0 21150.
2020 225 0 9999999.
2020 226 0 9999999.
2020 227 0 9999999.
2020 228 0 9999999.
2020 229 0 9999999.
2020 230 0 9999999.
2020 231 0 9999999.
Tell your pals the application is misbehaving. It should not return the last period if incomplete. It should stop at the last complete period. I wonder if the yearly average returns a data point with only a few days in January. Right now it returns a yearly average for 2020.
It is better for you if I don’t look at the data, because if I do I find all your mistakes.
I said that the difference was due to the graph being called on a different day and I was right. Archibald called the graph on the 18th of August and Willis on the 31st of July. That was the reason the graphs looked different. Spot on. Not bad for not knowing Omniweb.
Yes I did. I said the problem was that the last point was not a 27-days average as logic would dictate.
If you specify a 27-day average to Omniweb, all data points are ’27-day averages’ of the days on which there are data [which may vary from 0 to 27]. This you did not understand, so you did not do due diligence, but instead accused Willis of blunders and errors. Not good behavior. And not identifying the problem. That we all know by now so it does not do you any good to deny it.
“2020 223 0 2 25021.”
This data point does not contain a 27-day average. It contains only two points in the average. That is what the 2 in the fourth column means. This is the data point that made the difference between Archibald and Willis graphs. So I rightly identified the problem.
I was not the only one pointing to Willis that the difference was due to a single point. Two other commenters did. And Willis did mistake months for solar rotations, a point that you have studiously avoided while you make an issue of me not knowing how Omniweb returns the data. I don’t have to guess much what you would have said of me if I have made Willis blunders and errors. But hey, you are Dr. Bias.
Javier August 25, 2020 at 1:35 pm
You’ve made this accusation over and over without ever having the common courtesy to quote what it was I said. Please quote where I supposedly said that. I just went through the thread and can’t find anything like that.
w.
This data point does not contain a 27-day average. It contains only two points in the average
You learned about the ‘2 days’ here:
Leif Svalgaard August 23, 2020 at 3:41 pm
[…]It is the average of only a couple of days into the 27-days.
Now, lay off the broken record…
You know perfectly well that is what happened, and now you are pretending you don’t.
A full month, Willis? To the 31st of July? You didn’t even notice the graph was about solar rotations before going after David Archibald for a difference in the last point. And now you come all hypocritical pretending you knew it was about solar rotations, thinking you didn’t leave any evidence and you could get away impugning me. Shame on you, Willis.
That is just a confirmation of what I said earlier. The last point had an averaging problem.
Congrats on confirming my diagnosis of the problem.
That is just a confirmation of what I said earlier. The last point had an averaging problem.
No, it is not any such confirmation [regardless of your shamefully claiming that it is], because you did not know about the 2 days until I told you about them. And BTW, Omniweb does not have an averaging problem. Every data point is a 27-day average as advertised. No problem at all.
The essential point [that you miss] is that the average is always calculated using all available data within the 27-day Bartels rotation. Sometimes that is 27 daily values, sometimes that is 0 daily values and sometimes it is 2 daily values. The number of daily values can be anything between 0 and 27, both inclusive.
In short, you thought that the last data point was calculated from 17 daily values,as you had no idea about what the real issue was. So, your fraudulent claim that “I rightly identified the problem” is something you should shamefully retract. Perhaps wash your mouth out with soap.
I see you still don’t get it. Let’s start with the day numbers:
07/14/2020 is day number 196
08/10/2020 is day number 223
Last date allowed by Omniweb is 08/18/2020
correct range: 19631128 – 20200818
11/28/1963 is day number 333, and the first [lack of] value given is:
1963 333 0 27 9999999.
Which indicates that the day number is the first day of the rotation. This is confirmed because 19631128-19631225 returns just one rotation value.
Data is:
2020 169 0 27 39900.
2020 196 0 27 66177.
2020 223 0 2 25021.
So the last rotation started on 08/10/2020 and the data given by 08/18/2020 does not correspond to a full rotation. It only has 9 days for which only 2 data points are given. My diagnostic that there was probably an averaging problem with the last data point was correct despite being an hypothesis as I had not checked the data.
Your exchange with Ulric Lyons shows you didn’t get it (you still don’t).
You were wrong in that exchange and Ulric Lyons was right. The last full 27-day average ends on 09/08/2020. It is clear you didn’t understand what Omniweb was returning, and it is clear that if the last date is 08/18/2020 then it is not a full rotation. That day is day number 231 that does not fit the rotation dates series.
I gave the right solution to Willis confusion. You came later and got it wrong. In this thread I have demonstrated that you were wrong about the trend (David Archibald was right) in the plasma temperature data and you came late and got it wrong about Omniweb data (Ulric Lyons was right about it). No wonder you are so furious and make all kind of attacks on me. It is you who makes fraudulent claims all the time and try to disguise your shortcomings attacking others.
You are wasting your time and making me waste mine. Just recognize your mistakes (or not, they are evident) and move on. Willis attack on Archibald was wrong regardless of Archibald’s hypothesis shortcomings. His data was what was available from Omniweb. Willis tried to hide he did not understand the nature of the data. You guys will stop at nothing to attack people you disagree with. That is an awful behaviour, and we you are called on it you complain loudly accusing others of what you do.
I gave the right solution to Willis confusion.
As I have demonstrated, you did not, no matter how much you squirm.
Wasting your time? Easy solution: go away and stop polluting WUWT with your attacks.
I told you about the 2 day issue, but you still persisted in wrongly believing that OMNIweb was returning a 17-day average. Evidently, your purpose was to attack Willis as blundering. This is despicable behavior.
Even Archibald knew that if the solar wind had any influence on the Earth, one would have to consider the totality of the data [hence he showed the variation of several solar wind variables]. If anything, the temperature would be expected to have the least influence. In calculating long-term trends, obvious outliers would have to be omitted. Doing everything right, the conclusion stands that solar activity and solar wind were increasing from the 1960s to the mid 1990s and decreasing again to the present.
It seems that you are afraid of showing any [perhaps non-existing] evidence of understanding the scientific method. That speaks volumes about your credibility or lack thereof.
Oh . Please. Give it a rest. You are all nit picking whilst Rome is burning. Everyone is settled on the CO2 nonsense whereas we should be investigating if maybe it could be something else that is keeping in the heat in the oceans.
E.g. what is the influence of ppm oil contamination lying on top of the oceans on the capture of heat?
I had not checked the data.
Without checking the data, one cannot make meaningful statements about the data, so you did not make a correct diagnosis. Only after I pointed out the issue with the 2 days did you try to appropriate that as your own. In short: you did not understand the data or the issue, but only tried to take a swing at Willis.
It speaks volumes that you refuse to link to any evidence of your understanding of the scientific method.
You got it wrong. It is not about the 17th and 18th of August. You are the one that does not understand the issue even after being explained what it is. The same you cannot see an obvious trend even after being shown. No wonder you get so mad when being exposed.
As I said I will not attend any request that requires that I disclose my full name. I have a right to anonymity. I don’t care if you question my scientific credentials. Your approval is not required. I give ample proof of my climate knowledge. Keep at it as much as you want. It doesn’t bother me the least.
You got it wrong. It is not about the 17th and 18th of August
Nobody said anything about that. You claimed that the last point was a 17-day average, which is wrong. In short: you did not [perhaps still don’t] understand the issue. Only used the occasion to smear Willis.
And you are still not willing to show your stuff that you so proudly proclaimed would establish some credibility. You even post under a false name…
It is, of course, easier to be nasty when you don’t have the guts to tell us who you are.
No wonder you get so mad when being exposed
I do not get ‘mad’; instead I patiently for the benefit of the readership explain the science of the issue. For you? nah, you wouldn’t get it anyway. But there are other people here who deserve better.
You did when you told Ulric Lyons that the end date should be the:
That is wrong. It shows you did not understand the issue with the last point. Your claim is bogus.
Omniweb has a weird way of returning data. The last point starts on the 10th August and supposedly goes to the 18th, the last date that it allows to enter, but includes only two points. Going after Archibald for this is ridiculous.
It is curious how both Willis and you try to deny that you said what you said. That is a very devious behavior, more fit for a politician than a scientist. The more you insist, the worst it looks.
Omniweb has a weird way of returning data.
No, it is very reasonable. If you ask for 27-day averages, Omniweb always gives you 27-day averages [based of full Bartels rotations] of the days where there is data, even if only 2 or 1 or whatever that number happens to be.
It is always bad to base a trend on a rotation that only has a couple of days worth of data. That was the point Willis was making. What is bad form is to try and defend Archibald’s claim. In addition to turning the whole thing into relentless personal attacks and accusations [and hiding behind anonymity, not having the guts to say who you are] especially since you didn’t even go and look at the data.
That is not true for the last rotation (which is the cause of the discussion). The last rotation started on the 10th of August as I have showed you, and it will not end until the 5th of September. It will certainly have more than 2 days with data. Most probably it will have 27 days with data, or close to that number. The last time a rotation didn’t have 27 days of data was in 2003 and it had 26. So Omniweb returns an incomplete average for the last rotation until all the data for that rotation is added.
So you keep saying things that are not true (what a surprise). The whole thing is Omniweb’s fault for producing such an artifact. And Willis and you are being exposed and getting what you deserve for attacking and insulting Archibald, me and others like Ulric Lyons. You reap what you sow. Leave people alone and perhaps you will be left alone.
The last point starts on the 10th August and supposedly goes to the 18th, the last date that it allows to enter, but includes only two points. Going after Archibald for this is ridiculous.
It help to actually look at the data. Here they are with daily resolution:
The important point is that the various parameters end at different times. Some data providers deliver data on a monthly basis. To get a ‘level playing field’ it therefore makes as lot of sense to consider data ending times on a monthly basis. You can see that 31st July is such a good ending time as all parameters have data until then, and some don’t after that. What was perhaps confusing is that OMNIWeb in computing 27-averages actually goes beyond the ending day provided. This you didn’t notice [nor did Willis, actually], but Archibald was right in considering many solar wind parameters as if there is a solar influence it is likely not just a single parameter that does ‘the trick’ [and he went wrong in concentrating on the temperature [that BTW is 20-40% uncertain before about 1978].
Considering the complexity of all this, there is no reason to accuse people of blunders, errors, or worse. Shame on you for that [and for not having the guts to post openly]
You should tell that to Willis. He was quick to point that he couldn’t reproduce Archibald’s data without even trying. You were also quick to come to Willis defense despite knowing that there was nothing wrong with Archibald’s data.
I fundamentally disagree with David Archibald’s hypotheses, and there’s a lot to be criticized in them, but not the data.
He was quick to point that he couldn’t reproduce Archibald’s data without even trying
The issue was not to reproduce some sloppy work, but to point out that the serious drop at the end of the data was spurious. We all know by now that it indeed was spurious so Willis did a good job at first spotting this. As to Archibald’s claim, it was not worth taking serious and certainly now worth wasting 400 comments on, nor displaying so much malice and so many insults on… Instead of conducting such an extended smear campaign you should rather thank Willis for putting his finger on the salient point. Hopefully you have now run out of steam and venom so we can give credit where credit is due.
I fundamentally disagree with David Archibald’s hypotheses
In order to disagree there has to be some substance to disagree about. Nonsense like Archibald’s can only be dismissed out of hand.
Not spurious but butchered and badly averaged by Omniweb. That changed nothing in David Archibald’s article. The 45-year long-term trend in plasma temperature is unchanged by adding a 2-day average at the end. Willis made an issue of a red herring, fueled by his lack of knowledge about solar rotations, and you helped him with your stubbornness in refusing to recognize the trend in the data. The reality is that neither you nor Willis had a good understanding of why Omniweb was producing the chart that Archibald used, but that didn’t stop you both for charging against Archibald on the issue.
The venom and insults come out from your part when your shenanigans are exposed for all to see. That is what makes you so popular among the subjects of your vitriol.
If nonsense like Archibald’s can be dismissed out of hand why go after him for a red herring, a graph he didn’t make that is correct except for a point out of over 720?
The 45-year long-term trend in plasma temperature is unchanged by adding a 2-day average at the end. Willis made an issue of a red herring, fueled by his lack of knowledge about solar rotations,
It certainly changes Archibald’s eyeballed ‘trend’ and a meaningful trend should omit clear outliers as you have been told repeatedly.
Willis was correct in using 7/31 at the end point because the data providers to Omniweb often use deliver the data on a monthly basis. This has nothing to do with ‘lack of knowledge’, but everything to do with actually looking at the data [which you did not care to do]. You should thank Willis for attempting to ‘level the playing field’ in the manner he did.
The outliers was caused by coronal holes [where the wind speed is high resulting in shocks developing in interplanetary space (not on the sun) heating the wind anomalously]:
and
That you cannot understand this just shows you lack of knowledge of the physics of the solar wind. At least Archibald [in spite of all his faults] got it right that any possible influence of the sun depends on all the parameters [as he tries to make clear]. You, on the other hand, seem to be more interested in smear Willis than getting to the physics.
As I said, you should thank Willis for his effort in getting things right.
Archibald did not understand that the high plasma temperature does not comes from the sun. In fact, coronal holes [where high-speed solar wind comes from] are a lot colder than the rest of the sun. The high plasma temperature in space [near the Earth] is generated locally due to shock waves generated by the high-speed streams plowing into low-speed plasma and have nothing to do with radiation from the sun. So, he and you are barking up the wrong tree. Now, in your case that really doesn’t matter because you are only interested in smearing Willis, rather than getting to the truth of possible solar influence. At that from someone who doesn’t have the guts to comment under his real name.
Thank Willis for going after Archibald over an inconsequential red herring without a proper understanding of the issue? You should stop smoking that stuff. It makes you too funny.
Thank Willis for going after Archibald over an inconsequential red herring without a proper understanding of the issue
Yes, thank Willis. It was his understanding of the flaw in Archibald’s claim that laid it bare for all to see. His understanding was right on [trying to compare apples with apples by realizing that data providers often deliver on a monthly schedule and never on a Bartels rotation schedule, but you didn’t know that, did you?]. For you, understanding was not the issue, Willis-bashing was, and that from someone who does not the guts to put his name to his mouth.
I suppose we could look at hourly data:
I’ve never found use for Omniweb data before. But the way I see it you and Willis should thank me, because thanks to my insistence the issue has been clarified and Archibald exonerated. However I see you still don’t fully comprehend how Omniweb delivers its data because you are still making wrong claims. Omniweb data delivery does not follow calendar months as Willis assumed and you still say. Its plasma temperature data does not go to the end of July (as Willis thought), nor to the end of the solar rotation. It goes to the 12 of August, which is neither. I guess they run their own schedule at their convenience because the last date available for their query form is still the 18th of August.
I know, for example, that the HadCRUT database from UK MetOffice usually delivers a new value for a full calendar month towards the end of the following month, so they just released the July value. However sometimes it takes them longer. So I do know how some data providers work. What I have never seen before is that a request for a period average will return something else without saying.
Omniweb data delivery does not follow calendar months as Willis assumed and you still say.
But, the people who deliver data to Omniweb often do that, [as I showed an example of] so in order to get a situation where everybody has data going by months is the thing to do, as Willis rightly suggested. That Omniweb is a bit obscure about how the 27-day averages are computed is unfortunate, but my analysis and explanation to you clarified the situation and all is now well-understood.
But, as I said, your purpose was not understanding but Willis-bashing from a position of someone who does not have the guts to put a name to his attacks.
What I have never seen before is that a request for a period average will return something else without saying
That you have never seen it before only shows how limited is your understanding of data.
Omniweb does the logical thing [the only thing it can do]: if you ask for 27-day Bartels rotation data it gives you the average of all the data it has for each rotation. That can be[and often is] for less than 27 days, e.g.
Selected parameters:
1 Bartels rotation number
2 # of points in IMF averages
3 # of points in Plasma averag.
4 Scalar B, nT
5 SW Plasma Temperature, K
YEAR DOY HR 1 2 3 4 5
1982 143 0 2034 27 27 8.1 159478.
1982 170 0 2035 27 27 10.2 191523.
1982 197 0 2036 21 27 10.6 178293.
1982 224 0 2037 16 27 8.8 116047.
1982 251 0 2038 18 27 9.2 9999999.
1982 278 0 2039 20 8 8.0 9999999.
1982 305 0 2040 27 0 8.5 9999999.
1982 332 0 2041 27 0 10.0 9999999.
1982 359 0 2042 18 0 8.0 9999999.
very simple, very logical, and very necessary.
Only visual discrepancy seems to be that last point.. missing data ??
As you say, putting in a point to point line is sloppy to say the least !
It depends. Lines connecting the lows or the highs are very common in business and investing where they are also called trend lines. For example:
“Trendline Definition & Example
https://www.investopedia.com/terms/t/trendline.asp
A trendline is a line drawn over pivot highs or under pivot lows to show the prevailing direction of price. Trendlines are a visual representation of support and resistance in any time frame.”
I do not defend Archibald’s lines but I don’t think they are such an issue. As any trend line, it is something fictitious that doesn’t exist but acts as a visual aid to detect trends. And it is a common practice in other fields, like stock market analysis. Calling them bogus is ridiculous. All trend lines are bogus.
Willis, you need to make the end date the 18th August 2020.
Willis, you need to make the end date the 18th August 2020.
No, 16th August 2020, in order for the graph to show 27-day averages.
That still has the recent drop:
No,
The last full 27-day average excludes the 17th and 18h August and therefore does not drop.
My last link is to the 16th like you said.
The last full rotation shows an uptick in temperature
Selected parameters:
1 # of points in Plasma averag.
2 SW Plasma Temperature, K
YEAR DOY HR 1 2
2020 7 0 27 55501.
2020 34 0 27 67195.
2020 61 0 27 58825.
2020 88 0 27 65422.
2020 115 0 27 39312.
2020 142 0 27 44134.
2020 169 0 27 39900.
2020 196 0 27 66177. <—–
2020 223 0 2 25021.
But it is silly to based any conclusions on small amounts of data.
All solar wind data are correlated and the data show a general increase from the quiet sun in the 1960s through the active 1990s followed by a similar sized decline down to the recent quiet sun.
One thing I find terribly amusing about you Willis is that if other people don’t see the world exactly as you see it you get affronted, jump up and down on the spot and stamp your little feet. Our graphs do look different and there is no doubting your technical ability so I now accuse you of truncating the data set. When I switch Omniweb to ‘List Data’, the last data point is:
2020 223 25021
That is missing from your graph. It is “much below 5E+4” – as in half. I don’t think you did it deliberately to mislead. Omniweb is updated to March 18, 2020 and you most likely used a earlier date as your endpoint. But I am charitable like that, always trying to see the good in people.
That is missing from your graph.
As it should be as it is not a 27-day average.
And therefore introduces a false trend.
“David Archibald, I guess I’m the only person here who cares enough to try to replicate your data … sorry, I couldn’t. Here’s your Fig. 1.”
bingo
As always the number of real skeptics at WUWT remains fixed at 1.
Willis
what is it with these guys?
Do they forget that step 1 is check the guys data, and FOIA his ass if he wont share it
Ah so at least we can all agree that what Mann et al. in the climastrology community do is not science since they often claim that the data or models are proprietary and refuse to share them for replication. I’m glad you’re finally on board with step 1 in skepticism.
Dear heavens, the ignorance around here is bad. First Javier styles himself as knowing more about the sun than Leif Svaalgard does. Now you claim that Mosh is “finally on board” with sharing data …
Robert, I guess you weren’t around when Mosh and Steve McIntyre and Ross McKitrick and I and others were fighting the transparency fight. Mosh was with those in the forefront, I believe it was Mosh who made up the slogan “No data, no code, no science” … can’t find a more pithy statement of the requirements for modern science.
So your claim is not only wrong, it is hilariously and totally wrong. Guys, MOSH IS NOT WHO YOU THINK HE IS!!! He is a sharp scientist who thinks for himself and like me, doesn’t suffer fools gladly. My only complaint about him is his drive-by haiku style of commenting, where it seems you have to wring words out of him … but once he does explain it’s usually a valid and interesting point.
And yes, we disagree at times, sometimes passionately because we actually care … that’s why it’s called “science”, not “certainty”. But he’s always worth listening to.
Best regards,
w.
I am old now, but a long long time ago, one of my professors talked about the manipulation of data in various data subsets.
He showed where, with the proper adjustments via algorithms, that “ANY DATASET” could be manipulated to “STATE ANYTHING.”
He called the results of such manipulations: ” LS ” which was the acronym for: ” LIARS STATISTICS ” ……….
He is long dead, and i soon will be, but everything i have ever read about the alleged Global Warming phenomena appears to obviously be ” LS ” ….
Our little part of SE Australia has had an unusual winter. Cold, wet and snow down the unusually low levels. Much longer strings of days before seeing a break in this pattern.
Kp index is an indicator of disturbances in the Earth’s magnetic field caused by solar activity. Kp maxima usually lag about 3 years behind the sunspot cycles peaks.
http://www.vukcevic.co.uk/SSN-Kp.htm (normalised y scale)
but not same dates?
How can a blue line skirting highest points or lowest points of a graph be called an “average”? Maybe a trend of maxima, or a trend of minima?
By the way, the wild changes in the “solar plasma temperature” probably tell us more about the instrument used than about the solar plasma temperature.
Always worry. Panic when demanded. If you’re not in a constant state of anxious fear, then you aren’t supporting the cause of progress and change, and must therefore be a wrecker who stands on the wrong side of history.
Get in the pod, eat the bugs, consume product and get excited for more product.
In the long run the solar acitivity has the decisive role in the Earth’s temperature. The pause ended after 2014 but not because of anthropogenic reasons but because of the increase of shortwave radiation (the root cause is the change od low-level cloudiness). The solar insolation has been declining since 2000 but the shortwave radiation increased strongly after 2014 and it is the reason for the ending of the pause. Dr. Spencer publishes on mohtly basis the UAH temperature changes but he has missed the reason behind the trend.
Here is the link to my blog: https://www.climatexam.com/post/temperature-increase-since-2016-is-not-anthropogenic
The super El Nino 2015-16 finished the temperature pause and the temperature has stayed after 2019 about 0.4 °C higher than during the pause. The temperature change has increased by 45-50 % in a few years. Greenhouse gases cannot cause such a rapid change. Some climate researchers have suggested the delayed effect of greenhouse gases, but the real reason is the increased shortwave radiation from 2014 to 2019 total of 1.68 W/m2. The major change of 1.2 Wm-2 has happened after 2014. It means that the climate driver, which should be almost constant per the IPCC, has increased the temperature in 20 years by the same amount (1,66 W/m2 versus 1,68 W/m2) which took 270 years by carbon dioxide. This rapid temperature increase contests the IPCC’s climate models in two different ways.
The temperature of the GISTEMP dataset of NASA at the end of 2019 was 0.65 °C higher than in 2000. According to the science of the IPCC and NASA during 2000-2019 the temperature impact of greenhouse gases and aerosols was 0.3 °C and the effect of shortwave radiation was 0.8 °C. Totally it makes 1.1 °C, which is about 70 % higher than the observed temperature change. This result conflicts the climate models of the IPCC in two ways. The IPCC has assumed that the changes originating from the Sun are insignificant regarding the temperature changes. Another even more serious issue is that the IPCC’s climate models run away too hot. There is no such great error in my own climate model. Because the IPCC does not release any information about the shortwave radiation changes and its impacts, also media is silent about it. I have noticed that contrarian web pages are not aware about this issue.
My conclusion for years has been decadal climate has been driven by changes in cloud albedo. That has mainly been caused by changes in ocean circulation with especially the AMO/AMOC, ENSO and PDO oscillations.
The main summary being increased global sunshine hours and increased SW radiation warming the oceans because of less cloud albedo resulting in an overall decline in global reflected solar flux. Not only has this occurred this century to a less extent, but the biggest change occurred during the 1980’s and especially the 1990’s.
Antero, GISTEMP has been regularly tampered with to achieve the 0.65 °C increase.
Clouds are responsive to the tropical temperature, which is regulated by solar activity cycles.
Shortwave energy increased in 2014-2016 from the solar cycle maximum, and again in 2018/19 from fewer clouds, fewer clouds that resulted from less solar ocean warming under low TSI after the sun’s activity fell below the decadal warming threshold level in early 2016.
The IPCC models run hot because they feature a large fictitious AGW forcing component.
I am very well aware about the homogenization process that is the reason for the difference of 0.2 C degrees from 2000 to 2020 between UAH and GISTEMP. I used both temperatures to show that anyway the IPCC model showed about 70 % too high temperature increase from 2001 to 2020. I have shown in another study that SW change was the major driver of the super EL Nino peak temperatures of 1998-99 and 2015-16. This has not been discovered before in scientific studies.
According to my studies the TCS is 0.6 C degrees and not 1.8 C degrees of the IPCC.
I have shown in another study that SW change was the major driver of the super EL Nino peak temperatures of 1998-99 and 2015-16. This has not been discovered before in scientific studies.
Really? I discovered TSI was the major driver of both El Ninos
TSI is the total solar irradiation. What is the difference of TSI during those two super El Ninos? Have you published a scientific paper on this issue? What I see are only some images in your links.
Those examples are two different aspects of how TSI works during the solar cycle. I made a few posters for two sun-climate symposiums, one for the AGU and am working on a paper that will include all my work since 2014.
The subject definitely deserves more than a blog comment or two. I would rather present the material in the proper order, which I intend to do asap, and in there you will get a very specific answer to your insightful question.
We are very, very, doomed.
You talk about the 1970’s cooling and I know it warmed from there till 1998 then flatter but the graphs above seem to be shifted from those years. There may be a 5 to 10 year delay. I assume the climate averages of Earth take many years to fully reflect what is happening in the Sun just as they say the Earth is slow to fully react to CO2. It also takes a long time for the product of fusion to reach the surface of the Sun.
But I could be wrong, I’m not a “Climate Scientist” far less a “paid Climate Scientist with employment based on supporting AGW”.
I live in Texas – bring on the cold!!! It make take us from 100+ F days to…uh.. 100+ F days, sigh.
I would *really* enjoy the temps going down >>>In Texas<<< by 10 F or more. And more July and August rain while I am at it.
Meanwhile, there is plenty of cold beer available so I can survive.
I will believe these "The Sun is cooling" briefs when I actually see temperatures moving downward – until then it is just more wishful thinking. I want to see real non-manipulated data showing a long term trend, not a month's worth of unusual weather.
I am still a believer in the Earth is going to do whatever it wants to do, and the Sun is about the only known outside influence – it's a huge complex and chaotic system that drives climate. I acknowledge that none solar cosmic radiation *could* influence the climate, but I have to see it first. Most global warming (assuming that is even a meaningful term) is natural and anything man is doing through fossil fuel consumption is minor and just a blip on the overall timeline.
Meanwhile, I just read that global warming never happened before 1900 (LOL), that the Ice Age isn't real (LOL again), and that Methane is going to kill us all because there is too much in the atmosphere and in 100 years that will lead to more Catastrophic Warming. (How long does methane last in the atmosphere? 5 to 10 years maybe? But in 100 years it's going to kill us all so say their models)
I don't know why anyone cares, we all die in 2032 (or 2031?) anyway according to AOC. And it's all Trump's fault. (or is it 1/2 Bush's fault? I forget… I kind of miss blaming Bush for everything.)
I cannot imagine the circumstances where a rapid “plunge” (of 0.6C) in a reanalysis temperature estimate of the Southern Hemisphere from some obscure web site will grab much attention compared with the newsworthy features of California burning:
https://www.nbcnews.com/news/us-news/northern-california-wildfire-now-2nd-largest-ever-total-blazes-scorch-n1237778
The latter all due to Climate Change of course. The former simply a function of natural variation.
Another interesting aspect to the current circumstances is that cosmic rays are very close to a record level for the satellite era:
http://cosmicrays.oulu.fi
If you believe changes in wild fires are due to climate change then you will be relieved to know that it works in reverse. Global area burnt has been decreasing steadily over time for the last decades. See for example:
Earl, N., & Simmonds, I. (2018). Spatial and temporal variability and trends in 2001–2016 global fire activity. Journal of Geophysical Research: Atmospheres, 123(5), 2524-2536.
“There is a significant decline in 2001-2016 global fire activity, especially in the Northern Hemisphere and in Africa.”
More global warming should reduce global burnt area even more. In the meantime I suggest you folks invest in keeping fuel material low in forests if you don’t like wildfires.
Now, I’m far from an expert on the solar wind. That would be Leif. But as I said, I learn backwards, first looking at the data and then consulting the experts. Here’s an example of how I go about it.
Looking at the data, it seems to me like the connection of the temperature of the solar wind and the temperature of the surface of the sun is weak. The temperature of the solar wind goes from almost zero to 250,000K.
Meanwhile, during the sunspot cycle the output of the sun varies at the top of the atmosphere by about 1.5 W/m2 peak to peak around its average of 1,361 W/m2. Thats a peak-to-peak change of about a tenth of one percent …
Now, the sun has a temperature of about 5,778 Kelvin. So over the sunspot cycle it varies by about 58 Kelvin.
And over the sunspot cycle, per Figure 1, the temperature of the solar wind varies by on the order of ten THOUSAND Kelvin.
How are those two linked together? Particularly since the wind is often fifty times the temperature of the sun.
Next, the correlation of plasma temperature and sunspots is low. Plasma temperature lags sunspots by about three years. Here’s the scatterplot of the lagged correlation:
Although the solar wind does kinda vary with the sunspot cycle, the correlation is poor (0.4) and adjusted for autocorrelation, there’s no statistically significant relationship. In particular, when sunspot numbers are lowest, the solar wind speed are all over the map.
So I’m not seeing the connection where David Archibald says:
No matter what the solar wind has ever done or not done during the time we have records, the temperature of the sun hasn’t varied by more than about a tenth of a percent. And no, the Earth’s surface is NOT somehow “surely” commanded to follow that trivial change—a tenth of a percent change in anything is generally just lost in the noise.
Finally, solar activity has been steadily dropping since about 1980 … and the temperature has been increasing during that time. This is the exact opposite of what David and other “It’s the Sun, Stupid” folks claim. They say when solar activity drops, temperature drops … sorry, but the facts say otherwise.
Anyhow, that’s what I found out about the solar wind … so now I gotta ask Leif what I didn’t find out. Here are my questions after analyzing the data:
• Are there errors in what I said above?
• Is it the heliomagnetic field that is whipping the solar wind plasma up to those incredible temperatures?
• Is there a direct physical connection and relationship between solar temperatures and the solar wind?
• What is the reason for the 36-month lag between sunspots and the solar wind? Is it real or an artifact of a short record? If real, is it a result of the travel time for the solar wind spiraling out from the sun? (If so, then a change in the solar wind today reflects a change in the sun three years ago … but I digress)
My best to all, particularly the long-suffering and patient Leif.
w.
Are there errors in what I said above?
No, but it is more complicated
Is it the heliomagnetic field that is whipping the solar wind plasma up to those incredible temperatures?
No, the solar wind comes from the corona that have temperature in the millions. The wind temperature is generally decreasing with distance [it is expanding…]. It is a ‘kinetic’ temperature. If you a thermometer into the wind [in the shade] it would show only a few degrees above absolute zero [-270 C].
Is there a direct physical connection and relationship between solar temperatures and the solar wind?
No, the solar temperature [away from sunspots] is very constant [does not even vary with the solar cycle].
What is the reason for the 36-month lag between sunspots and the solar wind?
The corona is heated by sunspot magnetic fields, so the wind temperature will roughly follow the solar cycle but also depend a bit on the shape of the corona [the extent of its ‘streamers’] that move in latitude over the cycle so there is a statistical lag.
P.S. The exact mechanism for heating the corona is not really known; or rather: there are many different mechanisms in play, and we cannot agree on which is [ or are?] the dominant.
Thanks as always, my friend.
w.
Finally, solar activity has been steadily dropping since about 1980 … and the temperature has been increasing during that time. This is the exact opposite of what David and other “It’s the Sun, Stupid” folks claim. They say when solar activity drops, temperature drops … sorry, but the facts say otherwise.
Willis’ classic solar strawman argument. OK accountant, understand this – a person with $20K in annual fixed expenses is still gaining wealth if their income is halved from $60K. Solar activity operates similarly on the climate.
The question isn’t whether the sun’s activity has fallen since the 1980s, it’s whether solar activity was still high enough during this period for decadal-scale warming.
Solar activity hasn’t failed to warm the climate; Willis has failed to understand how and why.
As I have shown the warming after 2014 is due to the shortwave radiation increase despite of TSI decrease. Loeb et al. have found that it is due to the low-level cloud decrease. What is the reason for that is unknown.
Again, the reason is known – clouds are a response, not a driver.
Papers written where clouds are the drivers are greatly mistaken.
This is an example for the AMO that has to force the cloud feedback for it to maintain status quo. Without the cloud feedback the SST anomalies in the North Atlantic will reverse and lose its presence.
“Thus, the cloud feedback directly induces a surface radiative heat flux response that enhances the AMO-related SST variability over most of the North Atlantic. This enhanced SST variability supports SLP and surface wind responses that are sufficiently large to produce an anomalously positive TNA turbulent heat flux.”
Just one example that changes in ocean circulation force cloud feedback.
https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/2016GL068303
W
“No matter what the solar wind has ever done or not done during the time we have records, the temperature of the sun hasn’t varied by more than about a tenth of a percent. And no, the Earth’s surface is NOT somehow “surely” commanded to follow that trivial change—a tenth of a percent change in anything is generally just lost in the noise.”
You have ignored what the solar wind does to Earth’s surface temperature.
https://www.sciencedirect.com/science/article/abs/pii/S1364682616300360
Ulric Lyons August 24, 2020 at 1:11 pm
“Ignored”??? Osculate my fundament. I never heard of the damn paper as far as I know, and the amount of garbage on the subject is endless. No way I can keep track of it. Keep a civil tongue in your head. Perhaps you and your friends “ignore” things.
I don’t.
The paper you link to is another example of garbage. Of course, you didn’t link to the paper, just to the press release, indicating that you might not have even read it. I had to go get the paper from SciHub, it’s here. They claim a correlation between certain specific attributes of the solar wind and the North Atlantic, which is about 10% of the earth’s surface area.
There are a few problems with that. First among them is the lack of a Bonferroni correction. This is the correction for looking in lots of places. If you look in enough places you’ll find what looks like a correlation … but it’s likely just random chance.
How many places have they looked? Well, they claim to see the effect in a tenth of the globe. I see that they looked at various chunks of the other 90% to see what it contained. So there’s maybe ten places they’ve investigated.
But that pales beside the number of possible parameters of the solar wind and the interplanetary magnetic field. Here you go:
Fifty parameters, out of which they’ve picked four … presumably because that’s what correlates best with the surface temperature of the selected 10% of the earth. So … 50 solar wind choices times 10 choices of 10% of the surface area gives 500 possible combinations.
But wait, as they say on TV, there’s more. They only find the correlation in winter, one season out of four … so we’re up to 2,000 possible combinations.
And somewhere in there, you are pretty much guaranteed to find a correlation.
The Bonferroni correction adjusts the significance levels for the number of places you look. Apparently they never heard of it.
Next problem is that there’s no similarity in the periodograms (power spectra) between the solar wind and the and the North Atlantic. They do their best to disguise this. Here’s how.
Note the three solar variables all have power in the eleven-year range, just like the sunspot cycle. But there is no corresponding power at eleven years in the North Atlantic Oscillation.
But that’s not the disguised part. Check out the scales on the y axes …
Next, despite looking in all of those places, the correlation still wasn’t good. So they added two more variables to the mix, the interplanetary magnetic field variable Bz, and the “quasi-biennial oscillation” (QBO). The quasi-biennial oscillation (QBO) is a quasiperiodic switch of equatorial stratospheric winds between easterlies and westerlies.
Heres their abstract:
So their claim is that the North Atlantic winter temperature (but not the other seasons) is describable as some complex function of the form
T = f(GEF,Bz, QBO)
where GEF is the solar wind geo-effective electric field, Bz is a particular measure of the interplanetary magnetic field, and QBO is the quasi-biennial oscillation.
I can’t even imagine how many tunable parameters are in that equation … and I can’t count them, because they never explicitly spell the equation out. Another point against their analysis.
Finally, oceanic datasets are often highly autocorrelated, and this one is no exception. The long-term (1880-) sea surface temperature average dataset they used has a Hurst coefficient that is very, very high at 0.92.
This overlap hugely reduces the “effective N”, the equivalent number of actually independent datapoints. The full dataset they used has 1,804 months. But the autocorrelation is so great that the number of effective independent data points is under ten … meaning that all their statistical calculations need to be redone, as they have not dealt with autocorrelation or even mentioned it.
So once again, when I DO WHAT YOU DIDN’T DO, Ulric, when I actually go and get the paper and analyze what they did, I find that it is just another piece of statistical innumeracy masquerading as science.
Did I “ignore” this POS?
No, but I wish I had.
Next time YOU do the analysis, and YOU check the autocorrelation and the Hurst exponent, and YOU look at the number of choices and tunable parameters, and YOU see if they’ve used the Bonferroni correction BEFORE you come bother me with more of this bovine waste product … because I can assure you, having taken a long hard look at more “It’s The Sun, Stupid” papers than likely anyone you know … most of them are as bad as this one. They find some tiny effect on some specific part of the planet only in the wintertime, and specifically when the QBO is easterly, the moon is in the seventh house, Bz is decreasing, and Jupiter aligns with Mars … and they proudly declaim “SOLAR WINDS AFFECT THE GROWTH RATE OF MAN-IN-THE-MOON MARIGOLDS” … and folks like you swoon and recommend it to their friends.
Dial up the skepticism, folks, and do NOT depend on the modern joke called “peer review”. And if you cannot look at a paper yourself and identify the types of problems that I discussed above … you might consider not getting all enthusiastic about the paper …
Best to all, run the numbers yourself,
w.
Willis. Your paragraph which I had quoted flat ignores any solar wind wind effects and you refer instead to the small changes in solar irradiance.
So what if it’s only 10% of the planet, the AMO has global impacts and dominates interdecadal global temperature change. You should know that.
Most of the solar metrics are irrelevant, so your 2000 choices are a spoof. The clue you need is in the article title; correlations with the ‘solar wind speed’.
There will not be any 11 year power in the NAO or the AMO because the major lows in the solar wind shift in and out of phase with sunspot cycles every 2-4 solar cycles.
The sci-hub link won’t open here for some reason so I cannot comment on your autocorrelation point.
Your emotional ramblings about Jupiter and Mars and the Moon and Marigolds just shows that you’re short of real arguments against their findings of the inverse correlation between N Atlantic SST’s and the solar wind speed. You’re mocking the most important dynamic in whole climate system, it’s the reason for the AMO and Arctic always being warmer during centennial solar minima.
Ulric Lyons August 25, 2020 at 6:12 am Edit
Since the solar wind quite clearly follows the sunspot and solar irradiance cycles … so?
Perhaps I wasn’t clear. The issue has to do solely with the statistics, not with the question of any putative “global impacts” or “interdecadal temperature change”.
In fact, if you’d read the article you’d see that the main correlation they claim is with the GEF, NOT the SW speed. Nice try, though.
It also seems that you don’t understand the need for the Bonferroni Correction, which they’ve totally ignored. They’ve picked one season out of four, and they’ve picked one part of the planet out of ten. That alone, right there gives them forty different combinations – summer in the western South Pacific, spring in the Indian Ocean, they looked at them all.
And that means for their conclusions to be valid they need to find a p-value of 0.05 / 40 ≈ 0.001 … good luck finding that.
And that does NOT include the fact that they had 50 solar wind variables to choose from, out of which they chose the GEF and Bz … not counting the fact that they’ve tossed in the QBO.
And that in turn means that their statistics, even BEFORE considering autocorrelation, are already way wrong … and they did NOT consider autocorrelation.
Sorry, Javier, but their statistics are junk.
And yet, despite your handwaving, there is a clear 11-year component in the solar wind … but NOT in the NAO or the AMO. I gave you the graphic above. If your explanation were correct (it’s not) that would not be the case.
Right. In other words, you’ve put on this passionate defense of the paper without ever once reading it … why am I not surprised?
“Emotional ramblings”? I was just making fun of the fact that they’ve coopered up the holes in their theory by adding new variables. This kind of ad-hoc addition is generally a sign of serious problems with the underlying theory.
I gave you a host of real arguments. You’ve responded by handwaving.
Please give us the EXACT dates of your “centennial solar minima”, and we can check your once again uncited, unsupported, unspecified, totally vague claim.
And in any case, I greatly doubt that we have more than two such “centennial solar minima” in the period of record … which makes your “ALWAYS being warmer” claim kinda … well … on shaky ground.
But sure. When are the “centennial solar minima”? I’m not gonna guess, you’ll just tell me I’m wrong …
w.
Javier? That’s a good one. You no longer know who you are talking to.
Javier August 25, 2020 at 5:07 pm
You’re correct, Ulrich, after while all the clueless idiots just blend into one …
w.
I see you believe you can insult other commenters. What a despicable behavior as Leif would say. You are just loosing it.
Willis wrote:
“Since the solar wind quite clearly follows the sunspot and solar irradiance cycles … so?”
I have been through this with you at least ten times in the past and you keep ignoring the facts. The major lows in the solar wind definitely do not follow sunspot cycles.
W
“And yet, despite your handwaving, there is a clear 11-year component in the solar wind …”
You are doing the handwaving while ignoring the facts, as usual.
” In other words, you’ve put on this passionate defense of the paper without ever once reading it … why am I not surprised?”
Yes I have read the paper a while back thanks, it shows an inverse correlation between the “solar wind speed” and North Atlantic SST’s.
“You’re correct, Ulrich, after while all the clueless idiots just blend into one …”
Such personal attacks are invariably projections.
You have to use some intelligence and knowledge to realise why they use winter only. Here is a pertinent analogy, an AMO signal is apparent in UK sunshine hours in winter and spring, but ther’s nothing at all in summer and very little in Autumn.
https://www.metoffice.gov.uk/research/climate/maps-and-data/uk-temperature-rainfall-and-sunshine-time-series
Here is the issue. There were major lows in the solar wind speed at 1969 and 1979-80 at sunspot maximum, then they shifted to around a year past sunspot minimum in 1997 and 2008-09. That is why AMO anomalies shift phase with respect to sunspot cycles.
https://www.woodfortrees.org/graph/esrl-amo/from:1880/mean:13/plot/sidc-ssn/from:1880/normalise
In about 5 billion years when it uses up its hydrogen and enters its red giant phase. Until then, don’t worry, be happy.
Mosher
I have the weather for you in Jesus’ days.
It was hot.
https://breadonthewater.co.za/2020/07/30/the-weather-in-jesus-time/
Hi Henry
Earth was much simpler place in them days
http://www.vukcevic.co.uk/Earth-Magnetic-Field-Intensity.htm
W.E.
What is the reason for the 36-month lag between sunspots and the solar wind?
I have a questionable hypothesis, but Dr. S can elaborate is he is inclined to do so.
The solar wind is made of streams of plasma (particles) from the sun propagating out into space. The solar wind temperature as well as the K factor ( indicator of disturbances in the Earth’s magnetic field caused by solar activity) usually lag about 3 years behind the sunspot cycles peaks.
http://www.vukcevic.co.uk/SSN-Kp.htm (normalised y scale)
Most of solar flares (CMEs etc) ‘presumably’ erupt vertically (i.e. perpendicular to the solar ‘surface’). In the first half of the sunspot cycle most of the solar magnetic activity is further away from the solar equator
http://users.telenet.be/j.janssens/SC24web/Butterflyflares.png
and may well miss the Earth (taking the Ecliptic plane into account).
As the flare-producing sunspot groups move closer to the equator, the Earth will be more exposed. Abut 3-4 years after the SSN peak activity will decline by up to 50%, sufficiently enough to override effect of the near equator magnetic activity.
@David Archibald
Why are you cherry picking min/maxs on the plots? You pick one set of 3 minimum spikes and even suggest that this is a disciplined process occuring but on other plots you draw lines missing maximums because they “don’t fit your agenda?” Presenting data as such does not help the debate. It only polarises it.
I can see no reason why you should be using a time corrected running average on all this ‘spikey’ data and looking for patterns from there. Yes it looks like the sun surface is cooling – but isn’t that an obvious read across from the declining solar activity?
“What is evident in [fig. 4] is the change in regime from the Modern Warm Period that ended in 2006…”
_____________________
What evidence supports the claim that any so-called ‘Modern Warm Period’ ended in 2006? According to the UAH lower troposphere satellite temperature data set (TLT), the rate of warming between the start of the data, Dec 1978, through Dec 2006 was +0.15C/dec; a total warming of +0.42C.
Since Jan 2007 the rate of warming in UAH_TLT has been +0.34C/dec; a total warming of +0.47C in less than half the time of the period ended 2006. Yes, trends over a shorter period are more prone to natural variability in ENSO etc, but even so, the UAH_TLT temperature record hardly convincing evidence that any ‘Modern Warm Period’ ended in 2006.
Someone who writes “Modern Warm Period that ended in 2006…” is not being genuine. Shame on WUWT for promoting this drivel.
Shame on you for attacking WUWT for its diversity and willingness to allow different points of view. Voltaire is attributed the quote “I disapprove of what you say, but I will defend to the death your right to say it”. It wouldn’t do any bad to teach children and adults this cornerstone of freedom of speech.
Javier
What is interesting to note is that the ice in the arctic & greenland is still melting, but it seems for sure now that the heat is coming from the bottom:
https://wattsupwiththat.com/2020/08/22/arctic-ocean-moorings-shed-light-on-winter-sea-ice-loss/
The waters of the SH have not really warmed in the past 40 years. The light blue line in the wft plot shows ca. 0.1 K from 1979, which is really next to nothing?
I don’t even think that it is possible to measure all the waters of the SH oceans to that kind of accuracy. Note that that result also corresponds with my own results showing a warming of only 0.0017K/annum of 27 terrestrial stations in the SH compared to almost 0.024K/ annum in the NH (over the past 40 years)
Loydo will pretend to support diversity, until it diverts from his/her view. Typical of the hypocrisy of current neo-marxism.
Javier
“Shame on you for attacking WUWT for its diversity and willingness to allow different points of view.”
_____________________
Having a different point of view is well and good where the evidence available is ambiguous. Unfortunately that can’t be said of this claim made by the author here. Not a single global temperature data set supports his claim that recent global warming ended in 2006. It’s a nonsense to pretend otherwise.
Global temperatures have continued to rise since 2006 and, in the case of UAH_TLT at least (often the data set of choice at WUWT), at a rate faster than that seen over the long term up to that point.
That might be in part due to recent ENSO, but it’s still a 14 year period of temperature rise that the author of this article is willfully ignoring. Being entitled to a point of view doesn’t entiltle people to make up their own ‘facts’, surely?
He has a right to be wrong, but deciding that he is wrong and preventing him from publishing is censorship. The consensus decided that Alfred Wegener was wrong for 40 years. I am convinced that David Archibald is wrong, but convictions aren’t worth much in science.
You defend the disingenuous? Sleeping with dogs will give you fleas.
Perhaps. You appear to be flea ridden from sleeping with the climate alarmists.
So you know, the climate crisis is so old news… Nearly everybody moved over to the COVID crisis. Did they leave you behind?
Something strange happened on the sun around 1995, exactly as noticed in my report
(click on my name),
https://woodfortrees.org/plot/hadsst3gl/from:1979/to:2021/trend/plot/uah6/from:1979/to:2021/trend/plot/hadsst3sh/from:1979/to:2021/plot/hadsst3nh/from:1979/to:2021/plot/hadsst3sh/from:1979/to:2020/trend
causing the decrease in temperature of the SH oceans relative to the NH oceans
Any idea why? Anyone?
Correlation does not imply causation. The 65-year multidecadal oscillation reversed gears in the mid-90s. It is therefore another possibility, and one that is known to affect oceanic temperatures, since that is how AMO was discovered.
Javier
What is interesting to note is that the ice in the arctic & greenland is still melting, but it seems for sure now that the heat is coming from the bottom:
https://wattsupwiththat.com/2020/08/22/arctic-ocean-moorings-shed-light-on-winter-sea-ice-loss/
The waters of the SH have not really warmed in the past 40 years. The light blue line in the wft plot shows ca. 0.1 K from 1979, which is really next to nothing?
I don’t even think that it is possible to measure all the waters of the SH oceans to that kind of accuracy. Note that that result also corresponds with my own results showing a warming of only 0.0017K/annum of 27 terrestrial stations in the SH compared to almost 0.024K/ annum in the NH (over the past 40 years)
@vuk
My comment to Javier is applicable.
Looks to me my observations are consistent with the so-called magnetic stirrer effect.
(the earth’s inside is aligned with the sun’s inside)
Every ca. 1000 years or so.
Eddy cycle?
The article that you refer about Arctic sea ice being melted from warmth below is from Polyakov et al. This is preceded by an article in 2004 when Polyakov et al. showed that sea ice variability follows the 65-year oscillation over an underlying trend.
Polyakov, I. V., et al. “Variability of the intermediate Atlantic water of the Arctic Ocean over the last 100 years.” Journal of climate 17.23 (2004): 4485-4497.
“Here it is demonstrated through the analysis of a vast collection of previously unsynthesized observational data, that over the twentieth century Atlantic water variability was dominated by low-frequency oscillations (LFO) on time scales of 50–80 yr. Associated with this variability, the Atlantic water temperature record shows two warm periods in the 1930s–40s and in recent decades and two cold periods earlier in the century and in the 1960s–70s.”
I can only identify three factors affecting climate on a centennial scale:
1. The 65-year multidecadal oscillation responsible for the ~30 year trends, but neutral on the long term.
2. Long-term solar activity changes, responsible for most of the long-term trend.
3. GHGs, responsible for a part of the long-term trend.
The rest of factors appear not to contribute importantly to climate change.
Between 1975-2005 all three factors aligned in causing warming, producing the largely fictitious climate crisis.
Javier
Are you sure that the 65 year cycle is always 65 years?
I have read palaeo research that goes back a millenium or so and shows oceanic multidecadal cycles varying between 40-100 or so years.
For example:
https://agupubs.onlinelibrary.wiley.com/doi/epdf/10.1029/2004GL019932
(see fig 3 of this paper)
https://www.nature.com/articles/srep00609.pdf
https://openresearch-repository.anu.edu.au/bitstream/1885/67533/2/01_Heslop_Can_oceanic_paleothermometers_2011.pdf
I don’t think so. The evidence points to AMO having a different periodicity during the LIA, as in the Gray article you cite defends.
It is possible that the period of the multidecadal oscillation is affected by the long term solar activity level and changes from a grand minimum to the modern maximum.
Indeed – or it could just be that the solar periodic forcing of the AMO is weak, not strong, resulting in complexity of the emergent oscillation.
The south warmed when the AMO cooled from 1965. The southern SST’s are a fairer measure of long term temperature change.
https://woodfortrees.org/plot/hadsst3sh/from:1970/plot/hadsst3nh/from:1970
Bipolar seesaw?