“Skaters can only do this race every 10 or 11 years because that’s when the rivers freeze up,” Sirocko said. “I thought to myself, ‘There must be a reason for this,’ and it turns out there is.”
WASHINGTON – Scientists have long suspected that the Sun’s 11-year cycle influences climate of certain regions on Earth. Yet records of average, seasonal temperatures do not date back far enough to confirm any patterns. Now, armed with a unique proxy, an international team of researchers show that unusually cold winters in Central Europe are related to low solar activity – when sunspot numbers are minimal. The freezing of Germany’s largest river, the Rhine, is the key.
Although the Earth’s surface overall continues to warm, the new analysis has revealed a correlation between periods of low activity of the Sun and of some cooling – on a limited, regional scale in Central Europe, along the Rhine.
“The advantage with studying the Rhine is because it’s a very simple measurement,” said Frank Sirocko lead author of a paper on the study and professor of Sedimentology and Paleoclimatology at the Institute of Geosciences of Johannes Gutenberg University in Mainz, Germany. “Freezing is special in that it’s like an on-off mode. Either there is ice or there is no ice.”
From the early 19th through mid-20th centuries, riverboat men used the Rhine for cargo transport. And so docks along the river have annual records of when ice clogged the waterway and stymied shipping. The scientists used these easily-accessible documents, as well as additional historical accounts, to determine the number of freezing episodes since 1780.
Sirocko and his colleagues found that between 1780 and 1963, the Rhine froze in multiple places 14 different times. The sheer size of the river means it takes extremely cold temperatures to freeze over making freezing episodes a good proxy for very cold winters in the region, Sirocko said.
Mapping the freezing episodes against the solar activity’s 11-year cycle – a cycle of the Sun’s varying magnetic strength and thus total radiation output – Sirocko and his colleagues determined that ten of the fourteen freezes occurred during years around when the Sun had minimal sunspots. Using statistical methods, the scientists calculated that there is a 99 percent chance that extremely cold Central European winters and low solar activity are inherently linked.
“We provide, for the first time, statistically robust evidence that the succession of cold winters during the last 230 years in Central Europe has a common cause,” Sirocko said.
With the new paper, Sirocko and his colleagues have added to the research linking solar variability with climate, said Thomas Crowley, Director of the Scottish Alliance for Geoscience, Environment, and Society, who was not involved with the study.
“There is some suspension of belief in this link,” Crowley said, “and this study tilts the argument more towards thinking there really is something to this link. If you have more statistical evidence to support this explanation, one is more likely to say it’s true.”
The study, conducted by researchers at Johannes Gutenberg and the Institute for Atmospheric and Climate Science in Zurich, Switzerland, is set to be published August 25 in Geophysical Research Letters, a journal of the American Geophysical Union.
When sunspot numbers are down, the Sun emits less ultraviolet radiation. Less radiation means less heating of Earth’s atmosphere, which sparks a change in the circulation patterns of the two lowest atmospheric levels, the troposphere and stratosphere. Such changes lead to climatic phenomena such as the North Atlantic Oscillation, a pattern of atmospheric pressure variations that influences wind patterns in the North Atlantic and weather behavior in regions in and around Europe.
“Due to this indirect effect, the solar cycle does not impact hemispherically averaged temperatures, but only leads to regional temperature anomalies,” said Stephan Pfahl, a co-author of the study who is now at the Institute for Atmospheric and Climate Science in Zurich.
The authors show that this change in atmospheric circulation leads to cooling in parts of Central Europe but warming in other European countries, such as Iceland. So, sunspots don’t necessarily cool the entire globe – their cooling effect is more localized, Sirocko said.
In fact, studies have suggested that the extremely cold European winters of 2010 and 2011 were the result of the North Atlantic Oscillation, which Sirocko and his team now link to the low solar activity during that time.
The 2010 and 2011 European winters were so cold that they resulted in record lows for the month of November in certain countries. Some who dispute the occurrence of anthropogenic climate change argue that this two-year period shows that Earth’s climate is not getting any warmer. But climate is a complex system, Sirocko said. And a short-term, localized dip in temperatures only temporarily masks the effects of a warming world.
“Climate is not ruled by one variable,” said Sirocko. “In fact, it has [at least] five or six variables. Carbon dioxide is certainly one, but solar activity is also one.”
Moreover, the researchers also point out that, despite Central Europe’s prospect to suffer colder winters every 11 years or so, the average temperature of those winters is increasing and has been for the past three decades. As one piece of evidence of that warming, the Rhine River has not frozen over since 1963. Sirocko said such warming results, in part, from climate change.
To establish a more complete record of past temperature dips, the researchers are looking to other proxies, such as the spread of disease and migratory habits.
“Disease can be transported by insects and rats, but during a strong freezing year that is not likely,” said Sirocko. “Also, Romans used the Rhine to defend against the Germanics, but as soon as the river froze people could move across it. The freezing of the Rhine is very important on historical timescales.”
It wasn’t, however, the Rhine that first got Sirocko to thinking about the connection between freezing rivers and sunspot activity. In fact, it was a 125-mile ice-skating race he attended over 20 years ago in the Netherlands that sparked the scientist’s idea.
“Skaters can only do this race every 10 or 11 years because that’s when the rivers freeze up,” Sirocko said. “I thought to myself, ‘There must be a reason for this,’ and it turns out there is.”
Title:
“Solar influence on winter severity in central Europe”
Abstract:
The last two winters in central Europe were unusually cold in comparison to the years before. Meteorological data, mainly from the last 50 years, and modelling studies have suggested that both solar activity and El Niño strength may influence such central European winter coldness. To investigate the mechanisms behind this in a statistically robust way and to test which of the two factors was more important during the last 230 years back into the Little Ice Age, we use historical reports of freezing of the river Rhine. The historical data show that 10 of the 14 freeze years occurred close to sunspot minima and only one during a year of moderate El Niño. This solar influence is underpinned by corresponding atmospheric circulation anomalies in reanalysis data covering the
period 1871 to 2008. Accordingly, weak solar activity is empirically related to extremely cold winter conditions in Europe also on such long time scales. This relationship still holds today, however the average winter temperatures have been rising during the last decades.
Authors:
Frank Sirocko and Heiko Brunck: Institute of Geosciences, Johannes Gutenberg University Mainz;
Stephan Pfahl: Institute for Atmospheric and Climate Science, ETH Zurich, Switzerland.
==============================================================
I hope to have a copy of the paper soon – Anthony
UPDATE: Dr. Leif Svalgaard provides the paper, as did the AGU press agent Kate Ramsayer per my emailed request, along with a copyright admonishment. Thank you both. Figure 6a and 6b are interesting:
From the paper:
In agreement with the 20th Century Reanalysis central European temperature observations from the CRUTEM3 dataset [Brohan et al., 2006] from winters directly following a sunspot minimum are also significantly lower than the average temperature during the remaining winter seasons (Fig. 6a). The relation between cold winter conditions and sunspot activity is thus not specific to rivers alone (which could also be affected by a number of additional factors, for example warm water from the numerous powerplants constructed along the river). The strong variations of the time series in Fig. 6a, which are largely independent of the sunspot cycle, show the important role of internal, stochastic variability of the atmosphere for European winter temperatures. The relation shown above holds true only for central European temperatures. When the CRUTEM3 winter temperature data are averaged over the whole Northern Hemisphere, no relation to the solar minima is found.
This suggests a regional circulation pattern effect, as the authors state connected to figure 5a and 5b:
To identify the atmospheric circulation anomalies in the North Atlantic and European region associated with cold winters during solar minima, Fig. 5a shows the difference in the geopotential height fields at 500 hPa (Z500) between the winters directly following a year with a sunspot minimum and the remainder of the period 1871 to 2008, obtained from the 20th Century Reanalysis dataset [Compo et al., 1996]. A strong, statistically significant positive anomaly occurs over the eastern North Atlantic in the region of Iceland, while negative anomalies are found over the Iberian peninsula and over north-eastern Europe (the latter being not significant). These Z500 anomalies are associated with an enhanced northerly flow and cold air advection from the Arctic and Scandinavia
towards central Europe, leading to significantly negative temperature anomalies over England, France and western Germany (Fig. 5b). The centre of the cooling is in the region of southern England, the Benelux countries and western Germany down to middle Rhine area. Eastern and southern Germany are not effected as much as the above region. Accordingly, it is only the Rhine and possible some Dutch rivers that provide the possibility to reconstruct this specific temperature anomaly pattern, which corresponds to an anomalously negative NAO and a preference for blockings over the eastern North Atlantic.
Stephen Wilde says:
September 1, 2012 at 10:02 am
historical parameters that need large numbers of adjustments
Not so, reconciliation of the Group Sunspot Number and the Official SIDC sunspot number requires just TWO adjustments: http://www.leif.org/research/Reconciliation%20of%20Group%20&%20International%20SSNs.pdf
Stephen Wilde says:
September 1, 2012 at 10:24 am
We can already see that a wide range of climate linked parameters changed their trends at the same time as the sun moved from active cycle 23 to quiet cycle 24.
One swallow does not make a summer.
And the Sun has gone from active to quit many times. You think that the parameters changed every time?
Perhaps you are correct that only a few years is enough to falsify your ideas, but they are not enough to resolve the climate conundrum.
Studying the right variable to judge incoming energy would be a good starting point: maxima
I am sure that once I get the amplitiude of the ac curve – I will be able to determine the length of the cycle.
I think we will fiind the time of the cycle. Right now I can already estimate it is between ca ninety and one hundred years.
Of which half is warming and half is cooling time. Three times in that everything is as it always was if we ignore any other longer cycles that might also be relevant. H
Leif Svalgaard says:
September 1, 2012 at 9:30 am
Geoff Sharp says:
September 1, 2012 at 12:59 am
“Already the smoothed values for SC24 has reached [and passed] those for SC14″
cant you make it clearer so we can see some detail.
———————————————
Here is the version for people with poor eyesight:
http://www.leif.org/research/SC14-24-Smoothed-Adj.png
Note that the minima to start on clearly are 1902 and 2009. If you wish to deceive, you can choose another minimum year, e.g. 2008, but that is clearly not supported by the data.
Like I said the data you are using does not properly display reality. The last smoothed number is Feb 2012 and the values are temporary and will come down unless the next few months go mad. Interesting that you would use an obtuse form of measurement when you have been advocating all detailed data should be recorded. The unsmoothed data shows the real picture which at present shows SC24 closely matching the form of the GSN SC5. Joe D’Aleo also agrees.
http://icecap.us/index.php/go/joes-blog/unscientific_american_on_line_chat_on_heat_wave_to_cultists_blames_90s_this/
Leif Svalgaard says:
August 31, 2012 at 9:30 pm
Geoff Sharp says:
August 31, 2012 at 8:10 pm
You have employed a standard calibration for the pre Wolfer GSN values.
————————
The data seem consistent with a constant factor, but the calibration is done by carefully evaluating each observer: http://www.leif.org/research/What-is-Wrong-with-GSN.pdf
You will need to calibrate all observers using their telescope and drawings to determine their min spot size.
Considering that SC5 is mostly outside of 1700-1800, what Wolfer suggested does not apply to 1700-1800. In any event, we shall in due course resolve that question. Cosmic ray data for the 18th century suggests activity as high as in the 20th.
Obviously I was not referring to SC5 but criticizing your use of one factor to calibrate the GSN.
You correct for the Waldmeier jump by increasing SC14 by 20%
This is what You do. I think most of us would prefer to leave the pre 1945 figures alone and then remove the bogus Waldmeier factor after.
Wolf discovered a marvelous correlation between sunspots and the diurnal variation of the geomagnetic field. This was his greatest discovery.
Indeed a great discovery. But whether he applied it correctly in his reconstructions is under scrutiny and is challenged by Wolfer. I am not sure the Diurnal records go back far enough to calibrate 1700-1800?
Leif Svalgaard says:
August 31, 2012 at 9:48 pm
LSC is ill-conceived and poorly executed and has no value.
Over 400,000 views tends to suggest otherwise.
Geoff Sharp says:
September 2, 2012 at 4:07 pm
Here is the version for people with poor eyesight:
http://www.leif.org/research/SC14-24-Smoothed-Adj.png
One could argue that lining up on a whole year is inferior to lining up on a minimum defined by a fit to the whole minimum periods, like in http://www.leif.org/research/SC14-24-Smoothed-Adj-3.png
Like I said the data you are using does not properly display reality. The last smoothed number is Feb 2012 and the values are temporary and will come down unless the next few months go mad.
The smoothed sunspot number shows the reality that is the solar cycle without the day-to-day random fluctuations. If the next few months follow the present level, the smoothed number will stay where it is, just like SC14 did with just minor fluctuations over the rest of the maximum.
Geoff Sharp says:
September 2, 2012 at 4:23 pm
You will need to calibrate all observers using their telescope and drawings to determine their min spot size.
Nobody except Schabe and Wolf used a threshold so the minimum spot size is not relevant, and in any event the group counts stand on their own. But, we are actually digitizing Staudach’s drawings 1749-1806 and Schwabe’s drawings 1826-1867 to make sure. And ALL observers have their own independent calibration.
criticizing your use of one factor to calibrate the GSN
You are not paying attention. We found that the data shows that to first order there is a break in the calibration around 1885 [you cannot critisize the data for that]. The detailed calibration has a different factor for each observer.
I think most of us would prefer to leave the pre 1945 figures alone and then remove the bogus Waldmeier factor after.
The consensus in the SSN community is to correct the old data. The reason is simply that the current figures are used in operational programs [e.g. by the US Air Force – one of the sponsors of the workshops – to calculate atmospheric drag on satellites] so cannot be touched without causing unneeded problems. Since it doesn’t matter which way it is done, we oblige the users.
Indeed a great discovery. But whether he applied it correctly in his reconstructions is under scrutiny
No, it is not ‘under scruteny’ by anybody knowledgeable. On the contrary, the method is beeing accepted as the best hope for a successful re-assessment. We are aquiring 19th century data for that purpose. I just got the raw data from Prague 1839-1917.
and is challenged by Wolfer.
It was given up on by Wolfer, because he did know that the East component rather than the Declination was the controlling factor. This does not matter for periods of decades, but becomes evident on century time scales.
I am not sure the Diurnal records go back far enough to calibrate 1700-1800?
Then learn that the continous recording started in 1781 and we have scattered years of data from the 1740s and the 1760s.
Geoff Sharp says:
September 2, 2012 at 4:29 pm
“LSC is ill-conceived and poorly executed and has no value.”
Over 400,000 views tends to suggest otherwise
That is like suggesting that smoking is healthy because millions do it.
Geoff Sharp says:
September 2, 2012 at 4:23 pm
You will need to calibrate all observers using their telescope and drawings to determine their min spot size.
Not at all. Here is how Wolf is calibrated to Wolfer: http://www.leif.org/research/Wolf-Wolfer-Group-Calibration.png
Wolf and Wolfer used different telescopes and different techniques and did not use drawings, yet it is obvious by comparing their group counts for each year of their overlap [1876-1893] that with very high accuracy Wolfer recorded 65% more groups than Wolf. In the bottom left panel, Wolf’s count [with his small telescope – shown] is shown with blue symbols, Wolfer’s with pink. And Wolf’s multiplied by 1.653 with yellow. The pink and the yellow track each other closely. In a similar way all observers can be calibrated to other observers at the time and all calibrations combined to a solid composite [which matches the Zurich count closely]. The consensus is that the Group Sunspot number is flawed and will be abandoned. An example of the bad calibration used by H&S is that their factors for Wolf and Wolfer are identical to within 2%, which anybody by simple inspection of the data points in the lower left can see is wrong.
The SIDC value for August is once again higher than the NOAA discount value for the 17th straight month. The trend is rising since about 2009. Is the SIDC counting more or NOAA less?
http://tinyurl.com/2dg9u22/images/noaa_sidc_factor.png
Geoff Sharp says:
September 2, 2012 at 6:00 pm
The SIDC value for August is once again higher than the NOAA discount value for the 17th straight month. The trend is rising since about 2009. Is the SIDC counting more or NOAA less?
NOAA is intended for real-time use and they do not try to maintain long-term stability, so as I pointed out in a earlier comment, yes there is trend [ http://www.leif.org/research/SIDC-NOAA-k-png . But it is wrong to think that NOAA should have the same 0.6 k-factor as Wolfer other than by accident. So to compare a NOAA ‘discount value’ with SIDC as to which is higher does not make sense. what you can do is calculate the k-value for NOAA. For 1991-2001 the k-value was 0.662.
But, it is well-known in SSN-circles that SIDC undercounted [relative to NOAA and Sonne] in the years 2002-2009, this has now apparently been corrected.
Leif Svalgaard says:
September 2, 2012 at 5:27 pm
Nobody except Schabe and Wolf used a threshold so the minimum spot size is not relevant, and in any event the group counts stand on their own.
We have just seen that the Kremsmunters data employs a threshold. I find it hard to believe with the hundreds of observers involved along with the time frames and technology changes that all can be adjusted by one factor.
No, it is not ‘under scruteny’ by anybody knowledgeable. On the contrary, the method is beeing accepted as the best hope for a successful re-assessment. We are aquiring 19th century data for that purpose. I just got the raw data from Prague 1839-1917.
If its not under scrutiny the group is not performing their task adequately, is the team interested only in raising the early GSN values and not looking at the possibility of lowering the early SIDC values?
The raw data wont help with SC5. The Wolfer SC5 values and his methodology need to be fully explored.
Then learn that the continous recording started in 1781 and we have scattered years of data from the 1740s and the 1760s.
Like I said, not good enough to calibrate the 100 year period stated.
http://www.leif.org/research/SIDC-NOAA-k.png
Geoff Sharp says:
September 2, 2012 at 6:38 pm
We have just seen that the Kremsmunters data employs a threshold. I find it hard to believe with the hundreds of observers involved along with the time frames and technology changes that all can be adjusted by one factor.
No, there is no specific threshold ’employed’, simply an undercount because of only the larger spots being counted if they stood out, but not in any systematic way. And with the one factor, you are simply not paying attention. H&S has for each [well, most] of those hundreds of observers determined their individual k-factors. The problem comes in when they determined the k-factor for Wolf incorrectly. Since the Wolf data has a large weight in calculation, that skews the whole data set around 1885 and offsets the series, by about the difference between the correct Wolf factor and their incorrect one, hence leading to the discontinuity ~1885. So, in this way it really does come down to only one factor, because H&S have already taken all the differences into account.
If its not under scrutiny the group is not performing their task adequately, is the team interested only in raising the early GSN values and not looking at the possibility of lowering the early SIDC values?
The group has accepted my analysis as compelling [even Ken Schatten] so the scrutiny phase is past. The comparison with the geomagnetic data shows that it is GSN that is too low, so there is no possibility of lowering SIDC.
The raw data wont help with SC5. The Wolfer SC5 values and his methodology need to be fully explored.
As you could see that is the task for the 3rd SSN workshop to re-examine data 1750-1825.
Like I said, not good enough to calibrate the 100 year period stated.
But good enough to verify the calibration for the times when there is data.
Leif Svalgaard says:
September 2, 2012 at 7:58 pm
H&S has for each [well, most] of those hundreds of observers determined their individual k-factors. The problem comes in when they determined the k-factor for Wolf incorrectly. Since the Wolf data has a large weight in calculation, that skews the whole data set around 1885 and offsets the series, by about the difference between the correct Wolf factor and their incorrect one, hence leading to the discontinuity ~1885. So, in this way it really does come down to only one factor, because H&S have already taken all the differences into account.
So you are saying all observers before Wolf were calibrated via their k factor (GSN k factor) to match Wolf but Wolfer did not have the same matching test against Wolf? I would wonder how good the H&S GSN k factors are, considering the possible errors made so far??
The group has accepted my analysis as compelling [even Ken Schatten] so the scrutiny phase is past. The comparison with the geomagnetic data shows that it is GSN that is too low, so there is no possibility of lowering SIDC.
Sounds like a one sided agenda. There is obviously doubt raised on the Wolf recalibration before his count which the team is ignoring? SC3/4 are standing out as cycles that are possibly too high (Rz) when compared with the recalibrated GSN.
But good enough to verify the calibration for the times when there is data.
Wolfer has thrown doubt on this. If SC5 was adjusted from Wolf’s values, the previous values may also require more investigation.
Leif Svalgaard says:
September 2, 2012 at 6:45 pm
http://www.leif.org/research/SIDC-NOAA-k.png
Your graph is showing the same trend. No comment on this?
Geoff Sharp says:
September 2, 2012 at 9:24 pm
So you are saying all observers before Wolf were calibrated via their k factor (GSN k factor) to match Wolf but Wolfer did not have the same matching test against Wolf? I would wonder how good the H&S GSN k factors are, considering the possible errors made so far??
Observers were calibrated to the group count derived from the Greenwich Catalog 1875-1976, but before that they were calibrated largely to Wolf because he was the only one with a long series overlapping with Greenwich. However [and it is not known why – Ken can’t remember anymore] the calibration is wrong as I showed, so all older values become too low. This is the main error that H&S made. The SSN workshop is inclined to simply abandon the GSN number, like you do an old cars, when too many repairs are needed.
Sounds like a one sided agenda.
Truth only has one side. And science consists of presenting convincing and compelling arguments. Once that is done, agreement is quickly reached among reputable scientists.
http://www.leif.org/research/SSN/Hudson.pdf summarizes the discussion and
http://www.leif.org/research/SSN/Schatten.pdf is Ken Schatten’s take on it.
There is obviously doubt raised on the Wolf recalibration before his count which the team is ignoring?
What doubt? Wolfer abandoned Wolf’s method because he did not understand [actually could not know] that the parametr to use is the East component and not the Declination. The team understands completely the physics of the relationship and is not ignoring anything.
SC3/4 are standing out as cycles that are possibly too high (Rz) when compared with the recalibrated GSN.
Even though we can largely reconsile GSN and Rz, GSN is still on shaky ground because it is likely that some of H&S early calibrations are dubious. This will be addressed as the 3rd workshop, or rather: the GSN will be abandoned and the new data that H&S dug up will be incorporated in the new index we are building.
Wolfer has thrown doubt on this.
Wolfer did not throw doubt on anything, he simply did not understand what was going on.
If SC5 was adjusted from Wolf’s values
There were not, the values came from Kremsmunster. Wolf’s SC5 was as he left it when he died.
Your graph is showing the same trend. No comment on this?
I already commented: NOAA’s SSN is meant as a real-time, quick-and-dirty index for immediate use and does not pretend to maintain long-term stability. So any trend you might find has no special meaning and should not be overinterpreted.
Leif Svalgaard says:
September 2, 2012 at 10:17 pm
However [and it is not known why – Ken can’t remember anymore] the calibration is wrong as I showed, so all older values become too low. This is the main error that H&S made. The SSN workshop is inclined to simply abandon the GSN number, like you do an old cars, when too many repairs are needed.
If this is the case one wonders about the peer review process. If it was so wrong why was it not picked up, even with the questions raised the GSN is still a valuable piece of work ( The SC6 data in particular). I hope the rebuilding process is fully documented so we can all see the results and processes.
What doubt? Wolfer abandoned Wolf’s method because he did not understand [actually could not know] that the parametr to use is the East component and not the Declination. The team understands completely the physics of the relationship and is not ignoring anything.
The fact that Wolfer over rode some of Wolf’s figures for SC5 shows there was doubt in his mind, have you investigated his data and logic?
Wolfer did not throw doubt on anything, he simply did not understand what was going on.
I am talking about Wolfer using the Kremsmunster records, these are over riding some of the Wolf values which have the proxy record adjustment built in.
There were not, the values came from Kremsmunster. Wolf’s SC5 was as he left it when he died.
The 1902 Wolfer record looks to incorporate the adjusted values amongst the original Wolf values. Why try to confuse?
I already commented: NOAA’s SSN is meant as a real-time, quick-and-dirty index for immediate use and does not pretend to maintain long-term stability. So any trend you might find has no special meaning and should not be overinterpreted.
Another example of the data being no good because it is of no use to you. Do you have proper analysis that backs up your claim. You are claiming that the NOAA SSN is drifting towards an under counting position relative to their history? During the same interval I have noticed increased group splitting by Locarno which could also be a factor.
Geoff Sharp says:
September 3, 2012 at 12:21 am
If this is the case one wonders about the peer review process. If it was so wrong why was it not picked up,
The review process is not foolproof.
even with the questions raised the GSN is still a valuable piece of work
What is valuable is the additional data that H&S dug up.
I hope the rebuilding process is fully documented so we can all see the results and processes.
Everything we do is always fully documented.
The fact that Wolfer over rode some of Wolf’s figures for SC5 shows there was doubt in his mind, have you investigated his data and logic?>/i>
Wolfer did not override Wolf, as there was precious little to begin with. He added [the undercounted] Kremsmuenster to the data. The ‘logic’ was simple enough: Try to evaluate a k-factor using the better data from the 1820s and apply it backwards, but one can’t really overcome the undercounting [which as he states was especially problematic in the first decade of the data].
I am talking about Wolfer using the Kremsmunster records, these are over riding some of the Wolf values which have the proxy record adjustment built in.
see above
The 1902 Wolfer record looks to incorporate the adjusted values amongst the original Wolf values. Why try to confuse?
I don’t think he was trying to confuse. He did the best he could with crappy data.
Do you have proper analysis that backs up your claim.
I know how NOAA does their counting and what their data is meant for.
You are claiming that the NOAA SSN is drifting towards an under counting position relative to their history?
Rather returning to what it was earlier. And, as I said, it is likely that it is SIDC that have a problem during 2002-2009.
During the same interval I have noticed increased group splitting by Locarno which could also be a factor.
As Locarno is still mainly based on Sergio Cortesi’s observations there is no indication that they have changed. And ‘noticed’ is not good enough. How about doing a proper analysis?
The LSC/SIDC variance figures are in for August showing a 0.58 deviation. Some may say that is within tolerance of the 0.6 Wolfer factor, but in fact this deviation is much higher than it would seem. The variance factor mentioned only tracks groups that don’t make the grade (Wolf threshold) but Wolf not only didn’t count groups under a certain threshold where the biggest spot did not comply, he also did not count specks around groups that had a spot that did comply. Wolfer applied his conversion factor to all specks, so the 0.58 diversion is actually a long way from Wolfer’s overall 0.6 factor as the LSC counts specks around complying spots. This is more evidence that the Wolfer factor of 0.6 is not matching it during SC24.
Leif Svalgaard says:
September 3, 2012 at 4:53 am
Geoff Sharp says:
September 3, 2012 at 12:21 am
If this is the case one wonders about the peer review process. If it was so wrong why was it not picked up,
——————————
The review process is not foolproof.
Far from it perhaps. But digging deeper I see H&S clearly acknowledge that Wolf’s actual count is very short when looking at group counts, and yet they do not apply an appropriate k factor. They show two k factor columns per observer, not easy to fathom their logic.
even with the questions raised the GSN is still a valuable piece of work
————————————
What is valuable is the additional data that H&S dug up.
Yes, something that cannot be discarded.
I hope the rebuilding process is fully documented so we can all see the results and processes.
Everything we do is always fully documented.
The fact that Wolfer over rode some of Wolf’s figures for SC5 shows there was doubt in his mind, have you investigated his data and logic?
——————————-
Wolfer did not override Wolf, as there was precious little to begin with. He added [the undercounted] Kremsmuenster to the data. The ‘logic’ was simple enough: Try to evaluate a k-factor using the better data from the 1820s and apply it backwards, but one can’t really overcome the undercounting [which as he states was especially problematic in the first decade of the data].
Not so, there is plenty of Wolf values in the SC5 data that Wolfer finished with. You say under counted but looking at the data it is apparent only days with spots of Wolf type threshold were recorded by Derfflinger, the rest is entered as a zero and not referenced (meaning the coverage is much greater). I don’t have the full data but the Kremsmunster data could be meaningful. I will take it you have not looked into it with precision.
You give a lot of credence to the Diurnal record, but I have never seen a full record. Can you provide a record from 1780 to today that is compiled from the east component?
Leif Svalgaard says:
September 3, 2012 at 4:53 am
The 1902 Wolfer record looks to incorporate the adjusted values amongst the original Wolf values. Why try to confuse?
——————————–
I don’t think he was trying to confuse. He did the best he could with crappy data.
Well done, was meaning you are confusing and now you take it to a new level. I am not so sure the data is crappy but am sure its in your interest to persuade otherwise.
Do you have proper analysis that backs up your claim.
——————————–
I know how NOAA does their counting and what their data is meant for.
No evidence provided.
You are claiming that the NOAA SSN is drifting towards an under counting position relative to their history?
—————————
Rather returning to what it was earlier. And, as I said, it is likely that it is SIDC that have a problem during 2002-2009.
The Zurich/SIDC hierarchy has the runs on the board for regime change. Even today individual counters cant agree on the Waldmeier grouping rules. I see no evidence of this within NOAA?
During the same interval I have noticed increased group splitting by Locarno which could also be a factor.
——————————————–
As Locarno is still mainly based on Sergio Cortesi’s observations there is no indication that they have changed. And ‘noticed’ is not good enough. How about doing a proper analysis?
Yes but Sergio does not count everyday as you are aware. You have not done an analysis to show there is no deviation.
Never in my life have I seen two scientists argue about the wrong parameter for such a long time.
– have to laugh at you at all ur spots.
What is important is the energy coming thru.
Geoff Sharp says:
September 3, 2012 at 7:34 am
the LSC counts specks around complying spots.
And that is what is wrong with its execution. Wolf would rotate in his grave.
This is more evidence that the Wolfer factor of 0.6 is not matching it during SC24.
Wolfer’s 0.6 factor applies to HIM Alone using HIS method and HIS telescope with the seeing at HIS location and not to anybody else.
Geoff Sharp says:
September 3, 2012 at 8:00 am
Yes, something that cannot be discarded.
And it will not be. What is discarded is their faulty calibration
the rest is entered as a zero and not referenced (meaning the coverage is much greater).
No it means that there was no observation of the sun on that day. The primary purpose of the observations was the weather, not the sun.
I will take it you have not looked into it with precision.
I have looked at every day, but have not tried to calibrate yet.
You give a lot of credence to the Diurnal record, but I have never seen a full record.
Can you provide a record from 1780 to today that is compiled from the east component?
Actually could, because Wolf gives the values from 1781 to 1881 and I have my own list from 1882 to today. But it would be premature [and a waste] of time to do so, as we are collecting much more data than Wolf had and will do a proper job with all available data from the 18th and 19th centuries. From 1882 on, we have enough data to do a good job. Slide 29 of http://www.leif.org/research/Reconstruction%20of%20Sunspot%20Number.pdf shows such a reconstruction, which as you can see is even good enough the reveal the Waldmeier jump.
I am not so sure the data is crappy but am sure its in your interest to persuade otherwise.</i.
I'll tell it to your face: "you are full of shit". I am a careful scientist and working with several dozens other experts on sunspots to do the best job we can.
“I know how NOAA does their counting and what their data is meant for.”
No evidence provided.
My word is evidence.
Even today individual counters cant agree on the Waldmeier grouping rules.
They don’t need to as the k-factors take care of that nicely. The only thing needed is that each observer sticks to his own method [whatever it is].
I see no evidence of this within NOAA?
How many people do you know at NOAA? How many have you talked with about this? How many times have you visited them and discussed their procedures face to face?
Yes but Sergio does not count everyday as you are aware
So what? His k-factor for 2012 is 0.59, for the rest of the observers 0.61, and overall 0.61. For trained observers the k-factor is very stable from year to year. For Locarno it has been
2003 0.60
2004 0.60
2005 0.61
2006 0.61
2007 0.60
2008 0.62
2009 0.60
2010 0.59
2011 0.61
2012 0.61
The NOAA observers change often, typically every two years and NOAA’s k-factor never gets stable.
You have not done an analysis to show there is no deviation.
I never say anything without analysis.
HenryP says:
September 3, 2012 at 10:23 am
What is important is the energy coming thru.
What energy comes to the earth is well described by the spots. What comes through is mainly determined by the cloud cover [i.e. the Earth’s albedo], which varies a lot more than the energy from the sun. So, yes, the sun has little to do with it.
Leif Svalgaard says
What energy comes to the earth is well described by the spots. What comes through is mainly determined by the cloud cover [i.e. the Earth’s albedo],
Henry says
Naahh. Don’t think so. The amount of energy that comes thru, specifically that kind of energy that can heat the water of the big oceans (<0.5 um), is what determines your albedo; and yes, if less of that type of high energy comes thru, the atmosphere will get quicker saturated of water (g), due to the general cooling, meaning @ur momisugly lower humidity levels there will be some more clouds due to partial condensation. So it is a bit of a train/chain coming from the top down. I am pretty sure it starts with the change in ozone level on the top, which in turn was initiated by some minute change in the distribution curve of the TSI. I can see this from the deceleration curve of the warming (maxima) of the past 4 decades.
Did somebody already actually study the distribution curves of the solar constant at various times in the past – if this was ever done regularly?
HenryP says:
September 3, 2012 at 11:17 pm
I am pretty sure it starts with the change in ozone level on the top,
That you are pretty sure, does not make it so.