di Michele Moroni
Let's now go inside the details of Teodorani's report by revisiting some consideration already suggested by Matteo Leone and introducing new ones.
"By applying the technique of aperture photometry to the photographic sequence of Figure 1, it was possible to estimate an order of magnitude of the energy emitted by the light-phenomenon. This task has been accomplished by using the software Iris, after previous enhancement and resizing (image interpolation) operations have been carried out by using software Adobe Photoshop 5.5, by means of which small portions of a given photograph have been taken. Aperture photometry has been then applied by measuring the light-phenomenon on photo-frames by using concentric circles whose external radius matches exactly the radius of the light ball: in this way the apparent luminosity is directly calculated as a value in Erg/sec, then by multiplying by a factor 10^-7, the same parameter is obtained in Watts"
Beside the mistake in the description of the unit of measurement (the apparent luminosity should be expressed in Erg/s/cm^2 or in Watt/cm^2, much likely a careless mistake) it is unclear how Teodorani gets the values of the apparent luminosity, tabled at page 7. In fact the Iris Software only gives as output the counts number registered in the considered spectral band; this number is proportional to the apparent luminosity of the source. Closer attention by Teodorani is needed because the proportionality constant is not obviously known (technically is given from a procedure of convolution between the quantum efficiency and the spectral density curve of the source, all multiplied for another known constant). By being software for astronomy, Iris is only able to give the apparent magnitude related to the counts of the source (after a procedure of calibration with another known source). Due to these observations, it appears unclear how Teodorani could express a value in erg/s/cm^2 , writing it as being "directly calculated" by the software. Thus, I think the first passage needs clarification.
2) By accepting the correct
measurement of luminosity discussed in the previous passage,
it seems Teodorani assumes that the source emits the same
amount of lights in every direction (isotropic emission).
Basing on this assumption, he calculates the absolute
luminosity this time with the correct unit of measure. The
isotropic emission is a totally arbitrary assumption and I
don't think at the moment there is a way to prove it
(excluding simultaneous observations from two far
observers). In cases when the light emits with different
intensity in different directions Teodorani's evaluation of
the absolute luminosity would be completely nonsense. Since
the summer of 2002 some have speculated about the result of
a triangulation, but at the moment after 16 months nothing
has been published. Thus, I am sincerely starting to doubt
about its existence.
3) Teodorani evaluates the distance of the lights being 9 km, basing this result on a certain hypothesis that we assume as acceptable. He surprisingly decides to deduce the radius of the source from a comparison with the trees below ( 7) which are the ones of Heggsethogda hill, 4.5 km from the observing point of view. From the report it is totally unclear if the right proportional constant has been introduced. I wonder how come he did not calculate the dimension basing on the angular diameter and distance; the reasons of this choice are totally unknown.
4) Assuming the measurements
in the passage n.1 are correct and the absolute luminosity
isn't affected by the possible mistakes introduced in the
passage n.2, Teodorani evaluates the effective temperature
of the source assuming that it emits a blackbody. In case it
really emits as a blackbody this is the real temperature of
the emitting surface, at the opposite, they have not
correlation at all.
Thus, the photometric part is based on 4 chained passages, one certainly wrong and the other 3 ones certainly doubtful.
5) The choice to analyze the
spectrum recorded on a commercial film is not easily
understandable. In fact the film doesn't record the "real
spectrum" of the source because it is strongly modified by
the response of the support. Two different films can give
spectra with different features especially at the shortest
or longest wavelength.
"The sensitivity curve of the Kodak 100 Ektachrome film is not unknown to Teodorani et al. but the resulting spectrum, apart from some rough wavelength coincidences is different from the sensitivity curve mostly because of different ratios between the three peak amplitudes and because of a much higher amplitude of the single peaks".
This is a good reply because
avoids taking blame. He avoids mentioning the words "I used"
or "I did not use the correct response matrix of the
combination film/scanner". Thus, I certainly invite
Teodorani to tell this information without hiding it in
6) In the whole report Teodorani avoids to mention the error-bars associated to his measures; then it is impossible to know the precision of his measures. In the spectroscopic analysis not only does this important datum lack, but he refers to the unit of measure as "arbitrary unit" that prevents the reader to infer even a raw evaluation. This carelessness makes a comparison test between theoretical models and instrumental spectra extremely difficult to propose. Leaving Matteo Leone (and other researchers) almost unable to give the final prove to his assertions (if data lacks possible interpretations can be many!!). A so called "scientific report" without reference to the error bars would NEVER be accepted by a journal with a team of referees. Yet, judging the professionalism and the generosity of details of other Teodorani's reports he certainly knows this very well.
ABOUT TEODORANI'S REPORT
1b) The whole procedure, whose main-aim is proving that we are dealing with car- headlamps, assumes that the apparent brightness reported by Teodorani is correct while it is completely based on that questionable data (passage 1). If the measures of Teodorani were wrong, Leone's work would be partially compromised (though the excellent procedure developed would not be affected)
2b) As underlined by Iacopo Nicolosi and Nicoletta Ricchetti, Matteo Leone made a mistake evaluating some angular measures; the road on which Leone supposed the car is totally invisible to the observer. It is possible that the street is the one on Eggevollen Hill, 11 km far, but in this case, according to Leone's calculations, the absolute brightness would be much higher than the one expected by a couple of car-headlamps. I wouldn't pay much attention to this problem until we have a confirmation of Teodorani's evaluations of the apparent brightness. If that data was confirmed the strong case built by Matteo could start cracking.
3b) Matteo Leone showed, in qualitative way, how the continuous spectrum is extremely similar to the one expected by car headlamps convoluted by the typical response-matrix for a film... but doesn't make the comparison between the theoretic spectrum and the experimental one by a compatibility test (like a chi-squared test). Without a compatibility test Leone can certainly say "It is similar", Teodorani could reply "Not at all", and there is no way to go beyond this position.
Because of Teodorani's carelessness in evaluating the error-bars and giving the correct measure units the chi-squared test isn't easily usable, then I understand the reason because Leone did not propose it. If using other statistical tests was possible, I'm sure they will be proposed soon by Leone or Teodorani.
Note: After the publication of this article (Italian Version) Massimo Teodorani sent to my address a reply Email (private communication) declaring that not only he doesn't have any intention to reply my questions, but he warmly suggest to remove this article to avoid legal troubles. At the opposite I should thank Matteo Leone for his reply (private communication) declaring he is analyzing my considerations and he will produce a detailed integrative report soon.
(1) Teodorani, M. &
Nobili, G. (2002, October). EMBLA 2002: An Optical and
Ground Survey in Hessdalen. http://hessdalen.hiof.no/reports/EMBLA_2002_2.pdf
Copyright (2003) Michele Moroni