Skip to content

James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b

2022-07-15

Author: Hindemburg Melão Jr.

Reviewer: Scott Douglas Jacobsen

Numbering: Issue 30.B, Idea: Outliers & Outsiders (25)

Place of Publication: Langley, British Columbia, Canada

Title: In-Sight: Independent Interview-Based Journal

Web Domain: http://www.in-sightjournal.com

Individual Publication Date: July 15, 2022

Issue Publication Date: September 1, 2022

Name of Publisher: In-Sight Publishing

Frequency: Three Times Per Year

Words: 2,872

ISSN 2369-6885

Abstract

This article discusses the claims of the presence of water on exoplanet Wasp-96 b. There is a claim in some recent reportage in science news from the ESA-Webb news site. The author of this critical piece analyzes the claims about the allegations of the existence of water on the exoplanet Wasp-96 b. The author considers the article by ESA-Webb news more sensationalist than not.

Keywords: James Webb, James Webb Space Telescope, NASA, Wasp-96b, Water.

James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b

*Please see the footnotes and citation style listing after the article.*[1],[2]

*Original publication here: https://www.saturnov.org/news/james-webb, exception non-original submission permitted.*

It was recently reported that a study conducted with the James Webb space telescope has revealed the presence of water in the atmosphere of an exoplanet located 1150 light-years away. Indeed, there is some favorable evidence, but how credible is this claim? In this article we will examine how much fantasy and how much reality there is in this story.

Among the various recent news stories about the James Webb Space Telescope, this one, https://esawebb.org/news/weic2206/, in particular caught my attention for its sub-optimal (not to say poorly done) regression and the interpretation that was made of this graph:

Image credits: Hindemburg Melao Jr., Saturnov, Sigma Society.

There are many details that could be commented on, but there are two in particular that I would like to look at:

  1. among the several points indicated as signatures of the presence of water, only 1 is fairly clear (1.4 μm), and even in that case there is a large outlier nearby, which reduces the reliability in that signature. The points 1.9 μm and 1.1 μm are still reasonable, but the points 0.96 μm and 2.7 μm indicate nothing. In fact, the 2.7 μm point may indicate a counter-evidence.2. The regression model is apparently not parametric, perhaps some neural network was used for this, in which case a much better fit would be expected. The use of neural networks for this type of model should be used with great caution, because due to the large number of layers it is difficult to “see” and understand what is going on behind the fitting process.

When you use a model with 2 or 3 parameters, you know precisely what will change in the shape of the curve when you change each parameter up or down. But when you use a neural network, you are practically working with a “black box”, giving up understanding the underlying processes. In exchange, the use of this “black box” confers the “magical” power to achieve some operational advantages, among which is to get as good a fit as one wants, to corroborate any desired result (within certain more or less “plausible” limits). As a consequence, one can find what one wants, rather than finding the Truth. This is why I am generally against using neural nets, except in specific situations where one can compare them with results obtained by other methods, or as a fine-tuning step to improve a result that has already been determined by a more “cognizable” model, so to speak, as I have commented in my articles and my books.

I searched for the raw dataset on the Wasp-96 b light curve, but could not find it. On the James Webb website there is only data on the transits and the associated drops in brightness, but not on the absorption streaks at the different wavelengths. I would like to do my own modeling of the data, without the biases that apparently guided the intentions of the authors of this article, and calculate the probability that such results can be interpreted as a real signature of the presence of water. In any case, even without access to the data, it is still possible to make a panoramic analysis of what is on the graph, and the facts are not quite as the authors are suggesting.

It would be relatively simple and easy to make a better fitting model than the one used in this image taken from Webb’s site. In fact, it would be desirable to do at least two regressions, one without any a priori model, to try to capture the raw properties of what the data reveal to us, and another using a model on where the absorption streaks should be and what the absorption intensity is in each region, to make a more complete comparison of the entire morphology of the curves, rather than just comparing the positions of the ridges. This would provide a much more comprehensive view of the situation and help make a more reliable interpretation.

Both regressions would need to be robust, because there are many outliers that could shift the regression curve away from the region where it should be. Perhaps semi-robust models, such as Huber’s, would be more appropriate, because sometimes it is unclear whether or not a given point should be interpreted as an outlier, so it would be better to give each point a “credibility weight” or something, rather than entirely cutting off some points and leaving the rest entirely. Also, since there is little data in the sample, the uncertainty in determining which points are outliers is greater.

The fits near the crests of 1.4 and 1.1 are clearly bad, and visually one can already easily see that the curve of a good model should pass closer to the local central tendency of the point cloud in that part of the data, but it is passing far below.

One of the most “serious” problems is the distribution of the experimental points in the vicinity where the absorption line at 2.7 μm should be observed. This inconsistency at 2.7 μm may explain why the fit is (intentionally?) bad, because if they tried to do a better fit, the inconsistency would jump more clearly into view, so this degradation in fit quality may have been for the purpose of masking such inconsistency, to pretend that the problem in the hypothesis about the presence of water is not so serious.

A simple model with local adjustments of polynomials of order 3 every 7 points (or a little more), corrected at the extremities to connect smoothly, or something like that, would already provide a curve much more adherent to the points in the vicinity of the ridges, besides preserving the good fit in the other regions, making more evident the problem they tried to “disguise”. A Fourier series would also be an alternative to be considered, with the detail that the LPR fit needs fewer parameters in cases where there are long horizontal straight lines, or almost straight lines, because in these cases either the Fourier series needs a large number of parameters or it forces ripples that may not be representative of the reality where there are long straight segments.

In 2012, Liyun Su presented interesting work using “Local Polynomial Regression” (LPR), with clear advantages in many situations when one has a reasonable idea about the model a priori, but quite “dangerous” when one has no idea what to expect from the curve morphology. So it seems to me that the alternative I described above is able to meet well the criterion of being more adherent to the data, without substantially raising the risk of overfitting (as happens in some cases in which Liyun Su’s methods are applied). Therefore, the method I suggest in the previous paragraph would be preferable to both the method used by Su and the traditional Fourier series.

Wonsang You’s 2016 studies seem to me an interesting advance in the use of this tool, and Anna Derkacheva et. al.’s 2020 paper is decisively a good model for this purpose, with good local fits, low risk of overfitting, and when there are extensive and frequent càdlàgs in the time series, or substantial reductions in data density in certain regions, this type of method inhibits the emergence of large anomalies, as usually occurs with other methods. Although this is not a case of a time series, from a statistical point of view it can receive essentially the same approach, since each value on the x-axis has only 1 value on the y-axis, and every nth value is dependent on the (n-1)-th, which are some fundamental characteristics of time series.

Some of these studies are analyzed, refined and applied in my books IMCH and Two new rating systems. Here I will only give a brief introductory analysis, to clarify some of the most glaring problems.

The graph below shows an example taken from Anna Derkacheva’s article, in which some of the advantages of fitting a curve to a set of points with several undesirable symptoms are analyzed, and yet the fit looks very good and manages to avoid several typical errors that are often produced by other techniques:

Image credit: Anna Derkacheva.

Of course, this kind of model is useless for extrapolations, but it is extremely efficient for interpolations, deeply respecting the localities of the points along the whole considered spectrum.

However, none of this would be necessary to see the “error” in this case, which is quite obvious. By “glancing” at the graph on James Webb’s website one can already see that the 2.7 μm point probably contradicts the hypothesis of the presence of water, since in the vicinity of 2.7 μm there are empirical indications of a valley, but the hypothesis would require that there be a ridge. The use of more sophisticated models would only serve to formalize the detection of the problem and objectively demonstrate its presence.

Even if the uncertainties in this region (~2.7 μm) are large, due to the larger dispersion, this fact could not be neglected and the combined probability (in a Bayesian analysis) that the data set is a signature of the presence of water in the Wasp-96 b atmosphere becomes much more fragile when examining the situation with this approach. So it strikes me as a news story with symptoms of sensationalism, adopting a model with low adherence to the data, to force a “spectacular” interpretation in the eyes of the lay public, whose actual probability of being the correct interpretation is not that high, being far from conclusive.

One would need access to the raw data to calculate the probability that the null hypothesis (that the signal indicates the presence of water) should be ruled out. In the absence of the numerical values, looking at the graph, the most that can be done is a rough estimate. Within these limitations, the shape of the curve in the vicinity of the 1.4 μm point is very similar to that expected and the dispersion in this region is narrow. This is good for the author who argues that he has found evidence of water in the atmosphere of the exoplanet. It should also be noted that in the vicinity of the 1.1 μm point the fit is not as good, but still indicates a high probability. This is also a point in favor of the author of the article. But at the other points of the graph, the evidence is almost zero, and the big problem is that in the specific case of the 2.7 μm point, depending on the parameters of the model chosen to calculate the quality of fit, one can even have a counter-evidence.

A calculation that took this set of facts into account would perhaps indicate that there is more than a 90% probability that the data indicate the presence of water, which is indeed a strong indication, but at the same time the 10% probability that it is not an indication of the presence of water is something that could not be disregarded.

In particle physics and quantum mechanics, generally, when a new elementary particle is supposed to have been discovered, an evidence is considered “conclusive” when the accumulation of experimental data exceeds at least 99.73% probability that the observed signatures are of the particle sought. In some cases, only after the probability exceeds 99.99997% or 99.9999999% is it interpreted that the discovery has indeed been consummated. So a probability of “only” 90% is still a long way from what would substantiate such a strong statement, and should at the very least be viewed as over-optimism.

In any case, even if future research corroborates the presence of water in the atmosphere of this planet, some questions remain, including why they adopted such a poor model fit. The (probable) neural network itself used to generate the curve plotted in the graph published on J. Webb’s site is certainly versatile enough to allow a better fit, provided the author of the study wanted a better fit. This is reminiscent of those images of flying saucers, intentionally blurred to make it difficult to notice details that you don’t want to be noticed, lest you unmask something you wish to cover up. In my opinion, there is a lack of transparency, and the authors should at least try to justify why they preferred a bad setting. There are many situations where a “worse” fit is not only acceptable but also recommended to avoid overfitting, but this is clearly not the case. A brief discussion of this is done in my book about two new rating systems, in the part about a study by Rob Edwards to try to measure inflation in the rating. I also cover this topic in articles about investments and in some of my videos.

NASA, many other research centers and many universities have been doing this kind of cheap sensationalism for many years now. The news about life in the clouds of Venus, life on Mars, fossilized life in the Martian meteorite, life underground on Europa, evidence of a Dyson Sphere on another star, abnormal acceleration on Oumuamua, the imminent risk of explosion of Betelgeuse are just some examples of exaggerations and distortions, possibly intentional, because it is easier to get research funds this way. The sad reality is that science is just another enterprise at the service of the economy, advertising, politics, etc.

On the one hand, I think it is understandable that the pressure from the government and the people to justify investments in these projects forces researchers to make up the results, as well as needing to stitch together almost all astronomical research to force it to somehow connect to topics of popular interest, such as the search for alien life or saving the planet from an apocalyptic collision. I find this understandable, but I don’t agree. Surely there are more honorable ways to show the real importance of scientific results, without having to make them up to make them look “pretty” in the eyes of the general public. The ideal would be to educate the general public to enable people to appreciate and value real science, just as it is, instead of trivializing it to suit the taste of the lay public.

One of the problems of allowing science to be excessively contaminated by publicity is that some researchers may not distinguish between a serious and reliable study and one that is merely advertising, leading them to take as valid the results, whose errors end up spreading and contaminating other studies, in a chain reaction whose limits are lost sight of. This problem is much more frequent and more serious than one might think. In my book IMCH, I point out an alarming number of errors in data records from official sources, as well as several indications of fraud. Some authorities have been notified, but no action has been taken, indicating acquiescence with such frauds, or simple prevarication in some cases.

At this juncture, despite the various problems, the Financial Market is comparatively more transparent and more immune. I am not defending the Financial Market; on the contrary, I think that there are many problems of lack of ethics and a very worrying number of frauds not investigated, not supervised and not penalized in this sector. But there is a virtue that needs to be recognized: while in “science” one can play with theoretical models at will, without the errors being properly confronted with reality, in the Financial Market the models are confronted with reality all the time. This is why the use of accurate statistical tools to model the Financial Market is of crucial importance, so that one can maximize profits and minimize risks, with the results materializing quickly, impersonally and mercilessly. Errors and inaccuracies are punished, while success is rewarded in equal measure.

Nevertheless, in other areas, such as Psychometrics, Astrometry, Anthropometrics, Sociology, Cosmology, etc., mistakes are hardly punished at all. errors are hardly punished at all, and some errors are even awarded, with an unjustified recognition, supported by the lack of critical vision of the “peers” who analyze the cases under the same naive and incomplete prism of the authors of the awarded papers, leading to the exaltation of incorrect papers, simply because they fit better in the prevailing paradigms and dogmas, fitting better the collective beliefs of the majority of the academic community – in particular the beliefs of the members of the committees that determine who should receive the awards.

From this perspective, although the Financial Market can be cruel and bloodthirsty, it is also much more fair, more impartial, and produces results in accordance with the quality of the work developed. Not always the best works are the best rewarded in absolute terms, because the absolute gains depend on the percentage performances multiplied by the volumes under management, and as the vast majority of investors don’t have the basic knowledge to make good decisions about where to invest, it is natural that the vast majority of investors make bad choices, adopting shallow and irrational criteria. That is why the best investments do not necessarily produce the highest absolute profits, but they do produce the highest percentage profits, since these do not depend on the investors’ ability to choose, but only on the efficiency of the investment strategies themselves.

Footnotes

[1] Hindemburg Melão Jr. is the author of solutions to scientific and mathematical problems that have remained unsolved for decades or centuries, including improvements on works by 5 Nobel laureates, holder of a world record in longest announced checkmate in blindfold simultaneous chess games, registered in the Guinness Book 1998, author of the Sigma Test Extended and founder of some high IQ societies.

[2] Individual Publication Date: July 15, 2022: http://www.in-sightpublishing.com/webb; Full Issue Publication Date: September 1, 2022: https://in-sightjournal.com/insight-issues/.

Citations

American Medical Association (AMA): Melão H. James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b[Online]. July 2022; 30(B). Available from: http://www.in-sightpublishing.com/webb.

American Psychological Association (APA, 6th Edition, 2010): Hindemburg, M. (2022, July 15). James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b. Retrieved from http://www.in-sightpublishing.com/webb.

Brazilian National Standards (ABNT): HINDEMBURG, M. James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b. In-Sight: Independent Interview-Based Journal. 30.B, July. 2022. <http://www.in-sightpublishing.com/webb>.

Chicago/Turabian, Author-Date (16th Edition): Hindemburg, Melão. 2022. James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b.” In-Sight: Independent Interview-Based Journal. 30.B. http://www.in-sightpublishing.com/webb.

Chicago/Turabian, Humanities (16th Edition): Hindemburg, Melão “James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b.” In-Sight: Independent Interview-Based Journal. 30.B (July 2022). http://www.in-sightpublishing.com/webb.

Harvard: Hindemburg, M. 2022, ‘James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b’In-Sight: Independent Interview-Based Journal, vol. 30.B. Available from: <http://www.in-sightpublishing.com/webb>.

Harvard, Australian: Hindemburg, M. 2022, ‘James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b’In-Sight: Independent Interview-Based Journal, vol. 30.B., http://www.in-sightpublishing.com/webb.

Modern Language Association (MLA, 7th Edition, 2009):

Author: Hindemburg Melão Jr.

Reviewer: Scott Douglas Jacobsen

Numbering: Issue 30.B, Idea: Outliers & Outsiders (25)

Place of Publication: Langley, British Columbia, Canada

Title: In-Sight: Independent Interview-Based Journal

Web Domain: http://www.in-sightjournal.com

Individual Publication Date: July 15, 2022

Issue Publication Date: September 1, 2022

Name of Publisher: In-Sight Publishing

Frequency: Three Times Per Year

Words: 2,872

ISSN 2369-6885

Abstract

This article discusses the claims of the presence of water on exoplanet Wasp-96 b. There is a claim in some recent reportage in science news from the ESA-Webb news site. The author of this critical piece analyzes the claims about the allegations of the existence of water on the exoplanet Wasp-96 b. The author considers the article by ESA-Webb news more sensationalist than not.

Keywords: James Webb, James Webb Space Telescope, NASA, Wasp-96b, Water.

James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b

*Please see the footnotes and citation style listing after the interview.*[1],[2]

*Original publication here: https://www.saturnov.org/news/james-webb, exception non-original submission permitted.*

It was recently reported that a study conducted with the James Webb space telescope has revealed the presence of water in the atmosphere of an exoplanet located 1150 light-years away. Indeed, there is some favorable evidence, but how credible is this claim? In this article we will examine how much fantasy and how much reality there is in this story.

Among the various recent news stories about the James Webb Space Telescope, this one, https://esawebb.org/news/weic2206/, in particular caught my attention for its sub-optimal (not to say poorly done) regression and the interpretation that was made of this graph:

Image credits: Hindemburg Melao Jr., Saturnov, Sigma Society.

There are many details that could be commented on, but there are two in particular that I would like to look at:

  1. among the several points indicated as signatures of the presence of water, only 1 is fairly clear (1.4 μm), and even in that case there is a large outlier nearby, which reduces the reliability in that signature. The points 1.9 μm and 1.1 μm are still reasonable, but the points 0.96 μm and 2.7 μm indicate nothing. In fact, the 2.7 μm point may indicate a counter-evidence.2. The regression model is apparently not parametric, perhaps some neural network was used for this, in which case a much better fit would be expected. The use of neural networks for this type of model should be used with great caution, because due to the large number of layers it is difficult to “see” and understand what is going on behind the fitting process.

When you use a model with 2 or 3 parameters, you know precisely what will change in the shape of the curve when you change each parameter up or down. But when you use a neural network, you are practically working with a “black box”, giving up understanding the underlying processes. In exchange, the use of this “black box” confers the “magical” power to achieve some operational advantages, among which is to get as good a fit as one wants, to corroborate any desired result (within certain more or less “plausible” limits). As a consequence, one can find what one wants, rather than finding the Truth. This is why I am generally against using neural nets, except in specific situations where one can compare them with results obtained by other methods, or as a fine-tuning step to improve a result that has already been determined by a more “cognizable” model, so to speak, as I have commented in my articles and my books.

I searched for the raw dataset on the Wasp-96 b light curve, but could not find it. On the James Webb website there is only data on the transits and the associated drops in brightness, but not on the absorption streaks at the different wavelengths. I would like to do my own modeling of the data, without the biases that apparently guided the intentions of the authors of this article, and calculate the probability that such results can be interpreted as a real signature of the presence of water. In any case, even without access to the data, it is still possible to make a panoramic analysis of what is on the graph, and the facts are not quite as the authors are suggesting.

It would be relatively simple and easy to make a better fitting model than the one used in this image taken from Webb’s site. In fact, it would be desirable to do at least two regressions, one without any a priori model, to try to capture the raw properties of what the data reveal to us, and another using a model on where the absorption streaks should be and what the absorption intensity is in each region, to make a more complete comparison of the entire morphology of the curves, rather than just comparing the positions of the ridges. This would provide a much more comprehensive view of the situation and help make a more reliable interpretation.

Both regressions would need to be robust, because there are many outliers that could shift the regression curve away from the region where it should be. Perhaps semi-robust models, such as Huber’s, would be more appropriate, because sometimes it is unclear whether or not a given point should be interpreted as an outlier, so it would be better to give each point a “credibility weight” or something, rather than entirely cutting off some points and leaving the rest entirely. Also, since there is little data in the sample, the uncertainty in determining which points are outliers is greater.

The fits near the crests of 1.4 and 1.1 are clearly bad, and visually one can already easily see that the curve of a good model should pass closer to the local central tendency of the point cloud in that part of the data, but it is passing far below.

One of the most “serious” problems is the distribution of the experimental points in the vicinity where the absorption line at 2.7 μm should be observed. This inconsistency at 2.7 μm may explain why the fit is (intentionally?) bad, because if they tried to do a better fit, the inconsistency would jump more clearly into view, so this degradation in fit quality may have been for the purpose of masking such inconsistency, to pretend that the problem in the hypothesis about the presence of water is not so serious.

A simple model with local adjustments of polynomials of order 3 every 7 points (or a little more), corrected at the extremities to connect smoothly, or something like that, would already provide a curve much more adherent to the points in the vicinity of the ridges, besides preserving the good fit in the other regions, making more evident the problem they tried to “disguise”. A Fourier series would also be an alternative to be considered, with the detail that the LPR fit needs fewer parameters in cases where there are long horizontal straight lines, or almost straight lines, because in these cases either the Fourier series needs a large number of parameters or it forces ripples that may not be representative of the reality where there are long straight segments.

In 2012, Liyun Su presented interesting work using “Local Polynomial Regression” (LPR), with clear advantages in many situations when one has a reasonable idea about the model a priori, but quite “dangerous” when one has no idea what to expect from the curve morphology. So it seems to me that the alternative I described above is able to meet well the criterion of being more adherent to the data, without substantially raising the risk of overfitting (as happens in some cases in which Liyun Su’s methods are applied). Therefore, the method I suggest in the previous paragraph would be preferable to both the method used by Su and the traditional Fourier series.

Wonsang You’s 2016 studies seem to me an interesting advance in the use of this tool, and Anna Derkacheva et. al.’s 2020 paper is decisively a good model for this purpose, with good local fits, low risk of overfitting, and when there are extensive and frequent càdlàgs in the time series, or substantial reductions in data density in certain regions, this type of method inhibits the emergence of large anomalies, as usually occurs with other methods. Although this is not a case of a time series, from a statistical point of view it can receive essentially the same approach, since each value on the x-axis has only 1 value on the y-axis, and every nth value is dependent on the (n-1)-th, which are some fundamental characteristics of time series.

Some of these studies are analyzed, refined and applied in my books IMCH and Two new rating systems. Here I will only give a brief introductory analysis, to clarify some of the most glaring problems.

The graph below shows an example taken from Anna Derkacheva’s article, in which some of the advantages of fitting a curve to a set of points with several undesirable symptoms are analyzed, and yet the fit looks very good and manages to avoid several typical errors that are often produced by other techniques:

Image credit: Anna Derkacheva.

Of course, this kind of model is useless for extrapolations, but it is extremely efficient for interpolations, deeply respecting the localities of the points along the whole considered spectrum.

However, none of this would be necessary to see the “error” in this case, which is quite obvious. By “glancing” at the graph on James Webb’s website one can already see that the 2.7 μm point probably contradicts the hypothesis of the presence of water, since in the vicinity of 2.7 μm there are empirical indications of a valley, but the hypothesis would require that there be a ridge. The use of more sophisticated models would only serve to formalize the detection of the problem and objectively demonstrate its presence.

Even if the uncertainties in this region (~2.7 μm) are large, due to the larger dispersion, this fact could not be neglected and the combined probability (in a Bayesian analysis) that the data set is a signature of the presence of water in the Wasp-96 b atmosphere becomes much more fragile when examining the situation with this approach. So it strikes me as a news story with symptoms of sensationalism, adopting a model with low adherence to the data, to force a “spectacular” interpretation in the eyes of the lay public, whose actual probability of being the correct interpretation is not that high, being far from conclusive.

One would need access to the raw data to calculate the probability that the null hypothesis (that the signal indicates the presence of water) should be ruled out. In the absence of the numerical values, looking at the graph, the most that can be done is a rough estimate. Within these limitations, the shape of the curve in the vicinity of the 1.4 μm point is very similar to that expected and the dispersion in this region is narrow. This is good for the author who argues that he has found evidence of water in the atmosphere of the exoplanet. It should also be noted that in the vicinity of the 1.1 μm point the fit is not as good, but still indicates a high probability. This is also a point in favor of the author of the article. But at the other points of the graph, the evidence is almost zero, and the big problem is that in the specific case of the 2.7 μm point, depending on the parameters of the model chosen to calculate the quality of fit, one can even have a counter-evidence.

A calculation that took this set of facts into account would perhaps indicate that there is more than a 90% probability that the data indicate the presence of water, which is indeed a strong indication, but at the same time the 10% probability that it is not an indication of the presence of water is something that could not be disregarded.

In particle physics and quantum mechanics, generally, when a new elementary particle is supposed to have been discovered, an evidence is considered “conclusive” when the accumulation of experimental data exceeds at least 99.73% probability that the observed signatures are of the particle sought. In some cases, only after the probability exceeds 99.99997% or 99.9999999% is it interpreted that the discovery has indeed been consummated. So a probability of “only” 90% is still a long way from what would substantiate such a strong statement, and should at the very least be viewed as over-optimism.

In any case, even if future research corroborates the presence of water in the atmosphere of this planet, some questions remain, including why they adopted such a poor model fit. The (probable) neural network itself used to generate the curve plotted in the graph published on J. Webb’s site is certainly versatile enough to allow a better fit, provided the author of the study wanted a better fit. This is reminiscent of those images of flying saucers, intentionally blurred to make it difficult to notice details that you don’t want to be noticed, lest you unmask something you wish to cover up. In my opinion, there is a lack of transparency, and the authors should at least try to justify why they preferred a bad setting. There are many situations where a “worse” fit is not only acceptable but also recommended to avoid overfitting, but this is clearly not the case. A brief discussion of this is done in my book about two new rating systems, in the part about a study by Rob Edwards to try to measure inflation in the rating. I also cover this topic in articles about investments and in some of my videos.

NASA, many other research centers and many universities have been doing this kind of cheap sensationalism for many years now. The news about life in the clouds of Venus, life on Mars, fossilized life in the Martian meteorite, life underground on Europa, evidence of a Dyson Sphere on another star, abnormal acceleration on Oumuamua, the imminent risk of explosion of Betelgeuse are just some examples of exaggerations and distortions, possibly intentional, because it is easier to get research funds this way. The sad reality is that science is just another enterprise at the service of the economy, advertising, politics, etc.

On the one hand, I think it is understandable that the pressure from the government and the people to justify investments in these projects forces researchers to make up the results, as well as needing to stitch together almost all astronomical research to force it to somehow connect to topics of popular interest, such as the search for alien life or saving the planet from an apocalyptic collision. I find this understandable, but I don’t agree. Surely there are more honorable ways to show the real importance of scientific results, without having to make them up to make them look “pretty” in the eyes of the general public. The ideal would be to educate the general public to enable people to appreciate and value real science, just as it is, instead of trivializing it to suit the taste of the lay public.

One of the problems of allowing science to be excessively contaminated by publicity is that some researchers may not distinguish between a serious and reliable study and one that is merely advertising, leading them to take as valid the results, whose errors end up spreading and contaminating other studies, in a chain reaction whose limits are lost sight of. This problem is much more frequent and more serious than one might think. In my book IMCH, I point out an alarming number of errors in data records from official sources, as well as several indications of fraud. Some authorities have been notified, but no action has been taken, indicating acquiescence with such frauds, or simple prevarication in some cases.

At this juncture, despite the various problems, the Financial Market is comparatively more transparent and more immune. I am not defending the Financial Market; on the contrary, I think that there are many problems of lack of ethics and a very worrying number of frauds not investigated, not supervised and not penalized in this sector. But there is a virtue that needs to be recognized: while in “science” one can play with theoretical models at will, without the errors being properly confronted with reality, in the Financial Market the models are confronted with reality all the time. This is why the use of accurate statistical tools to model the Financial Market is of crucial importance, so that one can maximize profits and minimize risks, with the results materializing quickly, impersonally and mercilessly. Errors and inaccuracies are punished, while success is rewarded in equal measure.

Nevertheless, in other areas, such as Psychometrics, Astrometry, Anthropometrics, Sociology, Cosmology, etc., mistakes are hardly punished at all. errors are hardly punished at all, and some errors are even awarded, with an unjustified recognition, supported by the lack of critical vision of the “peers” who analyze the cases under the same naive and incomplete prism of the authors of the awarded papers, leading to the exaltation of incorrect papers, simply because they fit better in the prevailing paradigms and dogmas, fitting better the collective beliefs of the majority of the academic community – in particular the beliefs of the members of the committees that determine who should receive the awards.

From this perspective, although the Financial Market can be cruel and bloodthirsty, it is also much more fair, more impartial, and produces results in accordance with the quality of the work developed. Not always the best works are the best rewarded in absolute terms, because the absolute gains depend on the percentage performances multiplied by the volumes under management, and as the vast majority of investors don’t have the basic knowledge to make good decisions about where to invest, it is natural that the vast majority of investors make bad choices, adopting shallow and irrational criteria. That is why the best investments do not necessarily produce the highest absolute profits, but they do produce the highest percentage profits, since these do not depend on the investors’ ability to choose, but only on the efficiency of the investment strategies themselves.

Footnotes

[1] Hindemburg Melão Jr. is the author of solutions to scientific and mathematical problems that have remained unsolved for decades or centuries, including improvements on works by 5 Nobel laureates, holder of a world record in longest announced checkmate in blindfold simultaneous chess games, registered in the Guinness Book 1998, author of the Sigma Test Extended and founder of some high IQ societies.

[2] Individual Publication Date: July 15, 2022: http://www.in-sightpublishing.com/webb; Full Issue Publication Date: September 1, 2022: https://in-sightjournal.com/insight-issues/.

Citations

American Medical Association (AMA): Melão H. James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b[Online]. July 2022; 30(B). Available from: http://www.in-sightpublishing.com/webb.

American Psychological Association (APA, 6th Edition, 2010): Hindemburg, M. (2022, July 15). James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b. Retrieved from http://www.in-sightpublishing.com/webb.

Brazilian National Standards (ABNT): HINDEMBURG, M. James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b. In-Sight: Independent Interview-Based Journal. 30.B, July. 2022. <http://www.in-sightpublishing.com/webb>.

Chicago/Turabian, Author-Date (16th Edition): Hindemburg, Melão. 2022. James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b.” In-Sight: Independent Interview-Based Journal. 30.B. http://www.in-sightpublishing.com/webb.

Chicago/Turabian, Humanities (16th Edition): Hindemburg, Melão “James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b.” In-Sight: Independent Interview-Based Journal. 30.B (July 2022). http://www.in-sightpublishing.com/webb.

Harvard: Hindemburg, M. 2022, ‘James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b’In-Sight: Independent Interview-Based Journal, vol. 30.B. Available from: <http://www.in-sightpublishing.com/webb>.

Harvard, Australian: Hindemburg, M. 2022, ‘James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b’In-Sight: Independent Interview-Based Journal, vol. 30.B., http://www.in-sightpublishing.com/webb.

Modern Language Association (MLA, 7th Edition, 2009): Melão Hindemburg. “James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b.” In-Sight: Independent Interview-Based Journal 30.A(2022): July. 2022. Web. <http://www.in-sightpublishing.com/webb>.

Vancouver/ICMJE: Hindemburg M. James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b[Internet]. (2022, July 30(B). Available from: http://www.in-sightpublishing.com/webb.

License

In-Sight Publishing by Scott Douglas Jacobsen is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Based on a work at www.in-sightpublishing.com.

Copyright

© Hindemburg Melao Jr. and In-Sight Publishing 2012-Present. Unauthorized use and/or duplication of this material without express and written permission from this site’s author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Hindemburg Melao Jr., and In-Sight Publishing and In-Sight: Independent Interview-Based Journal with appropriate and specific direction to the original content. All interviewees and authors co-copyright their material and may disseminate for their independent purposes.

. “James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b.” In-Sight: Independent Interview-Based Journal 30.A(2022): July. 2022. Web. <http://www.in-sightpublishing.com/webb>.

Vancouver/ICMJE: Jacobsen S. James Webb Telescope: Alleged presence of water in the atmosphere of the Exoplanet Wasp-96 b[Internet]. (2022, July 30(B). Available from: http://www.in-sightpublishing.com/webb.

License

In-Sight Publishing by Scott Douglas Jacobsen is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Based on a work at www.in-sightpublishing.com.

Copyright

© Hindemburg Melao Jr. and In-Sight Publishing 2012-Present. Unauthorized use and/or duplication of this material without express and written permission from this site’s author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Hindemburg Melao Jr., and In-Sight Publishing and In-Sight: Independent Interview-Based Journal with appropriate and specific direction to the original content. All interviewees and authors co-copyright their material and may disseminate for their independent purposes.

Leave a Comment

Leave a Reply

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

WordPress.com Logo

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

Twitter picture

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

Facebook photo

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

Connecting to %s

%d bloggers like this: