Some years ago, on this blog, we had an animated correspondence on the subject of Scott’s ‘suicide’ – this was mostly centered on Professor Sienicki’s assertions in his paper ‘The Weather and its Role in Captain F. Scott and his Companions’ Deaths’, that Scott, having realised that there was no hope of him returning alive, decided on ‘a slow suicide’. Having reached this decision he was fearful that critics in Britain (who had previously slandered him and Lt.Royds, over the latter’s meteorological records from the ‘Discovery’ expedition), would pounce again, so, to gain public sympathy for himself and his own and his companions’ families, he and Bowers falsified their meteorological data and recorded abnormally low temperatures.
Professor Sienicki repeated these assertions in his book on the subject and was critical of other expert conclusions, mentioning particularly, Susan Solomon, the award winning atmospheric chemist.
As would be expected there was much argument about this conclusion and on this blog Bill Alp, an I.T. and software expert, persistently asked for details that would support Professor Sienicki’s data.
Now the American Meteorological Society has published a paper, which, without mentioning Sienicki,appears to refute his thesis.(http;//journals.ametsoc.org/doi/pdf/10.1175/BAMS-D-17-0013.1)
The paper shows that whilst several studies have focussed on the exceptionally cold conditions that Scott and his party suffered in the end of the Antarctic summer of 1912, both Scott and Amundsen experienced exceptional meteorological conditions – there were unusually WARM conditions in the interior of the Ross Ice Shelf and Amundsen’s party also experienced unusually high temperatures on the plateau as they approached the South Pole. At the same time Scott (well behind Amundsen) experienced these warmer than average temperatures on the Barrier. Scott also had higher-than-usual temperatures as his party descended the Beardmore Glacier. This warm period was followed by a colder than average temperature on the Ross Ice Shelf in early March. It is suggested, amongst other things, that the period of warmth may have lulled Scott into a sense of security before the temperatures dropped unusually and dramatically, sharply. There was no manipulation of the data — Scott’s cold weather observations were no more extreme than the high temperature observations.
Bill Alp has written the following:
It is great to see that a new research article, An Exceptional Summer during the South Pole Race of 1911-1912, has been published by the prestigious American Meteorological Society. It may be viewed at: http://journals.ametsoc.org/doi/pdf/10.1175/BAMS-D-17-0013.1
I like the article and recommend it to others because:
1. It provides an even-handed non-partisan analysis, including Amundsen’s meteorological observations, and it addresses the exceptionally warm period that Amundsen experienced in December 1911.
2. It does not propose any bizarre theories about Scott falsifying his temperature records in order to pave the way for his own suicide. Fogt et al simply show that most of the extreme temperature and pressure observations during this exceptional season are about two standard deviations above or below the mean value in the climatology model they used, called ERA-Int. Scott’s cold March temperature observations are no more extreme that Amundsen’s high December temperature observations. [Noting that observations of under-sledge temperatures were excluded because pooled cold air could cause a cold bias].
3. With a lifetime of experience in IT and software project management, I have wide personal experience in reviewing and probing the adequacy of testing that has been carried out in development of complex systems. Fogt’s article rings true for me.
4. The article has been through the peer review process of a respectable research institution.
So there we are. There will probably be more to follow
Isobel Williams – Author, Cruise Speaker and Lecturer 
“There will probably be more to follow.”
And so there shall be.
Fogt et al.’s temperatures are similar to Sienicki’s retrodicted temperatures (http://klimatolodzy.pl/pkp/pkp_21/Sienicki-pkp21.pdf as well as Figure 7.13 on page 300 of Sienicki’s book). Here is a link to a figure Sienicki made, in which he compares his old retrodictions (black squares) to Fogt’s Figure 4b (red circles) and Scott’s data (blue triangles): https://i.imgur.com/uz3T2Iw.gif
The remainder of the paper can be brought under distrust immediately, because Solomon is one of the coauthors. In her 2001 book, Solomon misrepresented the nature of Amundsen’s meteorological data to make it appear warmer than it actually was. With Sienicki’s permission, I am reproducing a section from subsection 4.2.2 of his book that explains this:
“Dr Solomon placed words in Captain Scott’s mouth by saying that “Scott noted that they were all feeling the cold,” and makes a junction that it was due to low temperatures. Captain Scott in his Jan. 14th journal entry was rather direct about the reason for feeling cold
Again we noticed the cold; at lunch to-day (Obs.: Lat. 89°20′53″ S.) all our feet were cold, but this was mainly due to the bald state of our finnesko.
Dr Solomon’s most glaring suggestion comes from the statement that the minimum temperature –19°F was “considerably colder than the conditions Amundsen had experienced a month earlier”. To support her observation, Dr Solomon produced Fig. 49 (see Fig. 4.10 above), a bizarre mixture of temperature data: minimum daily temperatures (Captain Scott), temporary temperatures (Captain Amundsen), average minimum daily temperature (modern data for the South Pole), and variability of minimum daily temperatures at the South Pole (modern data). In Dr Solomon’s presentation, the data depicted on this figure confirm her concealed thesis that Captain Scott’s team experienced significantly worse conditions than Captain Amundsen’s team, and thus Captain Scott was against the elements.
Dr Solomon gives no reference to where temperature data was taken. However, one can make an educated guess for all data sources, except for the temperature data attributed to Captain Amundsen’s expedition.
The data points, as well as the caption of Fig. 4.10, make reference to daily minimum temperature. However, Captain Amundsen never reported minimum temperature records, as the respective minimum temperature thermometers for unclear reasons were not taken by the expedition.462 Moreover, at the time of Dr Solomon’s book publication in 2001, the only source of Captain Amundsen’s expedition temperature data was available from his two volume published account of the journey, none of which was from his polar journey.463 In Tab. 4.2, I summarized the temporary temperature records found in Captain Amundsen’s printed account of the South Pole Journey in Dec. 10th through 20th, 1911. In the same table, I also added the temperature data attributed to Captain Amundsen by Dr Solomon. This data was obtained by careful digitization of Dr Solomon’s figure. Not even one of Dr Solomon’s data overlaps with Captain Amundsen’s actual record. Additionally, for Dec. 13th, 15th, 16th, 18th and 20th Captain Amundsen did not record temperatures in his printed record. I have no answer to the question of from what original source Dr Solomon assembled the temperature data for the just mentioned dates in December 1911.
It was Olav O. Bjaaland who, similar to Lt Bowers, kept meteorological records during the South Pole journey. His journal was not published until recently. Certainly Dr Solomon did not have access to it, and she did not make reference to Bjjaland’s diary. The respective temperature entries from Bjaaland’s journal are added to the previously mentioned Tab. 4.2. The gaps present in Captain Amundsen’s record are filled with false temperatures by Dr Solomon! Bjaaland’s temperatures are temporary records at different times of the day, and are not minimum temperatures as incorrectly suggested by Dr Solomon. As we are certain of Captain Amundsen’s and Bjaaland’s figures, one must conclude that Dr Solomon, not having original data sources, deliberately fabricated Captain Amundsen’s temperature figures to produce the impression that Captain Scott faced conditions at the South Pole much colder then Captain Amundsen did while there.
Provided that the most frequently recorded minimum temperature is during the night hours, as I have shown on Fig. 3.5, one, by looking at Captain Amundsen and Bjaaland’s entries in Tab. 4.2 can make an educated speculation that the respective minimum temperatures were substantially lower than the day temperatures, and in reality were closer to long-term averages obtained from modern data recorded at Amundsen-Scott Base and depicted by the solid line on Fig. 4.10. Therefore, both Captain Amundsen’s and Captain Scott’s minimum temperature data are uniformly scattered around long term averages of minimum temperature at the proximity of the South Pole. This is contrary to the false conclusion by Dr Solomon that the difference between Captain Amundsen and Captain Scott’s minimum temperatures was ‘more than ten degrees’.”
Well, that was an unexpected response! I was thinking you and Professor Sienicki, as men of science, would welcome with open arms any fresh research in your specialist area. Scientists, researchers and academics often like to use their knowledge, expertise and intellect, along with the research findings of others, to advance the frontier of human knowledge. You, however, have chosen to respond to the blog postings by reviewing Susan Solomon’s 2001 book.
Hmm… The phrase ‘fallacy of diversion’ comes to mind.
My main interest in the paper by Fogt, Jones, Solomon, Jones and Goergens is their finding that the Antarctic summer of 1911/12 was in fact an unusual season of extreme weather. Over the last 110 years, only 1911/12, 1925/26 and 1976/77 had mean summer pressures at McMurdo in excess of 1000 hPa. As the article states, “This ranks the McMurdo pressure during the summer of the South Pole races in the top three highest over the last 110 years, highlighting its exceptional character.” (page 2192).
It would be interesting to knowing if you and Professor Sienicki accept or dispute the finding that the Antarctic summer of 1911/12 was a season of exceptional weather.
If you dispute the finding, will either of you be publishing a counter-argument in a reputable journal?
If you accept the finding, are you able to refine the Professor’s ANN model to take on-board the learnings from Fogt’s article? How many seasonal data sets would be required to ‘train’ the ANN model to learn to classify the patterns associated with such rare seasons?
Hi Kristoffer
I would be very interested in knowing if you accept Fogt’s finding that the Antarctic summer of 1911/12 was a season of exceptional weather.
Bill
Dear Professor Sienicki and Kristoffer
I was looking forward to a polite, objective chat about the relative merits and limitation of your methodology and Fogt’s. But it is difficult to make a start without knowing if you accept or disagree with the finding that the Antarctic summer of 1911/12 was a season of exceptional weather.
It is disappointing that Professor Sienicki has not responded about the substance of Ryan Fogt’s paper ‘An Exceptional Summer During the South Pole Race of 1911/12’, published by the American Meteorological Society in October 2017. I have asked three times via this blog whether he accepts the findings of Fogt’s paper. In the absence of any response from the Professor, I will move forward on the assumption that his answer would be ‘yes’.
It appears in light of Fogt’s paper that Professor Sienicki faces some difficulties if he wishes to defend certain aspects his 2010 paper ‘The Weather and its Role in Captain Robert F. Scott and his Companions’ Deaths.’
1. Sienicki’s methodology involves ‘training’ his ANN model to learn to classify patterns in meteorological data recorded at McMurdo Base (inputs to the model), and from those inputs to calculate corresponding data for other specific points on the Ross Ice Shelf (in particular the Schwerdtfeger Automated Weather Station at -79.904°, 169.97°). The ‘training’ is achieved by processing a sufficient number of datasets to cover every possible meteorological pattern that will occur in the time period to which the ANN model is to be applied. The ANN model will not recognise and will not correctly interpret patterns that have not been encountered during training. Fogt’s assessment is that there were only three Antarctic summers in the last 110 years (1911/12, 1925/26 and 1976/77) where summer mean pressures exceeding 1,000 hPa. (Fogt, p. 2192) – hence his classification of 1911/12 as an exceptional summer. As I understand the Professor’s methodology, training datasets would need to include McMurdo and Schwerdtfeger for at least 1925/26 and 1976/77, before the ANN model could be seen as being adequately trained to predict / retrodict 1911/12 temperatures at Schwerdtfeger. However, Sienicki’s 2010 paper has been constrained by non-availability of AWS data prior to 1985. This non-availability of training data from these exceptional Antarctic summers raises concerns about the legitimacy of his approach to assessing the Polar Party’s temperature data (in 1912 or any other exceptional Antarctic summer). It is an ‘elephant in the room’
2. An observation that comes to mind is that Sienicki’s ANN-based scientific analysis has not produced any conclusions about the Terra Nova expedition that could not have been produced by comparing the Polar Party’s temperature data against simple arithmetic means of AWS temperature data. By having temperature as its single meteorological variable, Professor Sienicki’s ANN model is somewhat simplistic. Fogt’s paper shows that mean pressure is an important meteorological variable in modelling and understanding Antarctic weather events such as the 4-day warm wet spell experienced by The Polar Party from 5 December 1911, which Sienicki sidesteps with “Research into this question is needed” (Sienicki p. 7). Further, by making altitude adjustments, Fogt is able to include data from the polar plateau, something that Sienicki’s ANN model cannot do in its current state of maturity.
3. Sienicki uses two proxies in applying his ANN abstraction to Antarctic topography: (1) McMurdo AWS is used as a proxy for Wind Vane Hill (at Cape Evans where Simpson collected his meteorological data in 1911/12) and (2) Schwerdtfeger AWS is used as a single proxy for all the Polar Party’s locations between 27 February 1912 and 19 March 1912. In 2015, I raised questions with Professor Sienicki about his evidence for the validity of proxy (1) and was fobbed off. However I note Fogt’s statement “While pressures can be considered fairly uniform, local temperatures are well known to be highly variable around Ross Island and on parts of the Ross Ice Shelf, owing to, for example, katabatic and foehn winds and variations in sea ice cover” (Fogt, p. 2191). Scott reported temperature variations of up to 4 or 5 degrees Fahrenheit between the Cape Evans hut and the top of Wind Vane Hill (Scott, p.187, 192). I do not know what data exists about temperature variations in 1911/12 between the Cape Evans hut and the location which became McMurdo AWS many years later, but Sienicki treats any net difference between Wind Vane Hill observations and McMurdo AWS’s location as being insignificant. Given the scientific rigour applied in other areas of Sienicki’s paper, I see the lack of rigour applied to proxy (1) as another ‘elephant in the room’.
4. Fogt wrote about the cold spell experienced by Scott in late February and March 1912: “… although the daily averaged temperatures at this time were unusual, they were not record cold, and they do not fall below the two standard deviation threshold based on the ERA-Interim climatology” (Fogt, p. 2197). There is no suggestion by Fogt that Scott and Bowers falsified their temperature data from 27 February 1912 onwards. The ‘Results and Discussion’ section in Sienicki’s paper should be amended accordingly.
5. The final challenge that comes to mind is to identify a pathway for Sienicki’s ANN-model to be refined in order to take on-board Fogt’s work, so that further advances can be made in scientific modelling and analysis of Antarctic weather. Fogt’s paper does not take wind velocity & direction or cloud & sunshine into account as meteorological variables, so perhaps an ANN model that had input neurons for near surface temperature, atmospheric pressure, wind velocity & direction and cloud & sunshine could lead to significant advances in scientific understanding of Antarctic weather. Just a suggestion. With a more sophisticated model incorporating additional variables across a longer time span, it would be interesting to see if Sienicki’s pre-disposition to ‘mirrored similarity’ holds true for exceptional years.
References
Fogt, R.L. (2017) An Exceptional Summer During the South Pole Race of 1911/12 published by the American Meteorological Society in October 2017: https://journals.ametsoc.org/doi/abs/10.1175/BAMS-D-17-0013.1
Scott, R.F (2011). Scott’s Last Expedition Ware, England: Wordsworth Editions Ltd
Sienicki, K. (2010), The Weather and its Role in Captain Robert F. Scott and his Companions’ Deaths Podkowa Leśna, Poland: https://arxiv.org/pdf/1011.1272
Hello,
Sienicki’s comments on Fogt et al. (and Fogt et al.’s response to them) have now been published in BAMS:
https://journals.ametsoc.org/doi/pdf/10.1175/BAMS-D-17-0282.1
https://journals.ametsoc.org/doi/pdf/10.1175/BAMS-D-18-0088.1
He has written a summary for you:
1. Fogt et al. (2017) did not provide evidence or analysis that the daily temperatures for the statistical analysis on which they base their conclusions are independent and identically distributed (i.i.d.), and that those are normally (Gaussian) distributed. Fogt et al. assumption of normally (Gaussian) distributed data, is inappropriate according to textbook statistical analysis. Most of Fogt et al. conclusions are based on this false assumption.
2. Fogt et al. falsely argue that four data points well approximate a 24-h dataset.
3. Fogt et al. use computer generated temperature data. These data were never compared to the real data from weather stations along Scott’s route.
4. In their analysis, Fogt et al. did not use all of Scott’s temperature data and limited their analysis up to 3 March, though in Figs. 4b and 4c the depicted data end on 4/5 March. Thus, the most crucial set of temperature data recorded in Scott’s journal (Solomon 2001; Sienicki 2016) was omitted (see Fig. 2), and the comparison to a previous publication of one of the contributors is ineffective.
5. Fogt et al did not explain why the global El Niño and SAM “decide” to strike only Scott’s party and not the Cape Evans instruments?
In summary, Fogt et al. produced a cargo cult science article by comparing arbitrary computer generated data with Scott’s data that had been cut in half. Computer generated data (with 42 miles resolution) were not compared to data that were simultaneously recorded in Antarctica, and therefore cannot serve as reference points. No error bars are given and neither is the scientific method followed. 42 miles are equivalent to 7 or more of Scott’s daily marches which translates into (see Fig. 1) a 1C per day gradient. Thus, ±7C difference in temperatures recorded by Scott. Because 1F=5/9C, thus 7C translates into about ±14F. The error bar ±14F. Look at Fogt et al. plots and you find that ±14F blurs everything and no sensible observation is possible.
In addition, I have written this brief response to your 5 points:
1: The ANN was used to measure temperatures, not atmospheric pressures. Therefore, your whole point is moot.
2: Again, the ANN was used to measure temperatures, not atmospheric pressures, so your whole point is moot. In any case, the comment about the ANN’s state of maturity is incorrect, since it is simply a matter of feeding it AWS data from the polar plateau.
3: You were not “fobbed off”. You were given a quote from Sienicki’s book that dealt with the subject, which I will draw your attention to again: https://isobelpwilliams.com/2015/06/08/scotts-suicide/#comment-1759
4: Fogt may not have made any suggestion about Scott and Bowers, but I fail to see why Sienicki’s 2010 Arxiv paper needs to be revised on that basis.
5: I completely fail to see how these are related.