Eight Minutes Old

Two weeks ago my professor and me were interviewed by the Deutschlandfunk, which is a big German Radio Station. The topic, of course, was the new study concerning Dark Matter. The (German) broadcast can be found online as a Podcast and in text. This week we got another Interview request, now from NDR Info. As my professor and the other Co-Author from Bonn are at a conference in the US right now, I was the only one to be interview. You can imagine this made me a bit nervous, but it turned out to be fine.

The broadcast (again in German, sorry), will be aired today between 9 and 10 pm in the NDR Info show “Logo”. It will also be available as a podcast (starting at 39:30 into the file). I am curious what it will be like.

For Monday evening I was asked to talk about the Modified Newtonian Dynamics at our institute’s Astro Club. That’s a regular (voluntary) meeting for the younger physics students*, organized by several very committed tutors. They give easy to understand (and usually German) talks and, if the weather is fine, also observe with the institute’s telescope.

After that I hope to find more time for my work again. And for some posts on black and white infrared photography (film, of course), as I have already shown you two of my infrared photos in the post about faked UFO photos. As an introduction, here is another one. It shows the Sydney Harbor Bridge.

*: Of course older students are welcome, too.

Tags: Astroclub, deutschlandfunk, Harbour Bridge, MOND, podcast, Radio, Rollei Infrared, Sydney, Astronomers, Astronomy, Australia, Bonn, Canon EOS 50E, Dark Matter, EF 24mm f/2.8, Infared Photograpy, News, Science, Students, Study, University of Bonn

Finally I can tell you why I did not have much time to blog in the past month. Besides my own paper (have no news on this yet)  I wrote a popular science article together with my professor. That article will be published in the August Edition of Spektrum der Wissenschaft, which is the German version of the Scientific American.

Furthermore, I participated in another, very important research project resulting in a probably controversial, but scientificaly very important paper. Today it got the first place of the astrophysics preprintserver listing ArXiv: “Local Group Tests of Concordance Cosmology: Towards a new paradigm for structure formation?“. The study is about a test of the predictions of the cosmological standard-model including the dark-matter hypothesis on cosmologically small scales: In the range of the Milky Way satellite galaxies. The clear result is, that the standard cosmological model fails to agree with the observations. In total, the study lists five problems, ranging from the fact that satellite galaxies with different dark matter masses unexpectedly have the same luminosity in their center, to the phase-space distribution of the Milky Way satellite galaxies.  Each one is a serious problem for a scientific theory, but taken together the situation looks catastrophic. But, I think, it is good for science in general, because that’s the way science works: A theory makes predictions, which we use to test it. If the theory fails it is falsified and we need another, better description of the world. So it looks like we need an alternative to Dark Matter, which is great because we can expect some new fascinating results from this field of research in the next years.

First author of this study is my Professor (Pavel Kroupa) from the University of Bonn in Germany. Scientists from Germany, Austria, France, Italy and Australia collaborated for this research. And I am one of them. The part of work I did was about the disc of Milky Way Satellite Galaxies, and was large enough to give me place number five in the author list. I already mentioned the disc of satellites in my post about Tidal Dwarf Galaxies, but plan to present my new results in a post soon. So stay tuned.

UPDATE: The article is published in the August issue of Spektrum der Wissenschaft, and is available as a free pdf for download, too. You can find more information on this in this post.

Tags: Scientific American, Spektrum der Wissenschaft, Test, Astronomy, Cosmology, Dark Matter, Galaxies, Galaxy Formation, Milky Way, Paper, Physics, Press Release, Research Paper, Satellite Galaxies, Science, Scientific Method, Scientific Paper, Study, Tidal Dwarf Galaxies, University of Bonn

It is done. I have handed in my diploma-thesis. My work is done, now I only have to wait.

I do not want to write more about my project until it is published. But  you might get an impression from the word cloud (created by wordle.net*) below: Something about galaxy-scale model calculations with particles.  Once the paper is out you will get first-hand information.

So, now I have to write a paper about my work. And get my PhD-studies organized, and a trip planned. I did not blog this before, as it was not sure if it works out but now it is: I will be in Australia in December. I will visit the Australian National University (ANU) in Canberra. The Mount Stromlo Observatory, to be precise.  To get in touch with observers and give a talk there (I think it is on December 11). That will be cool, I never left Europe before.

*: Showing the 100 most-often used words, the bigger, the more often they appear in the text. It is not perfect as I simply copied the .pdf, leading to cut-off words (e.g. ar, pe, ep).

Tags: Astronomers, Astronomy, Diploma Thesis, Galaxies, Physicists, Physics, Students, Study, University of Bonn

These are the stories that are one of the reasons I enjoy astronomy: Three undergraduate (!) students, from Leiden University, discovered an extrasolar planet. And the best: they were not even looking for it. Their research project was to create a search algorithm, looking for light fluctuations in databases. Initially it wasn’t even planet to use their algorithm on the OGLE* database. But they had time left and within the brightness data of 15700 stars found one star which became fainter for two hours every 2.5 days. This was a strong hint at a planet passing by in front of the star, with an orbital period of those 2.5 days. When the planet is in front of the star, he shades some of the light. Because of the huge distances in space, the star (nor the planet) can not be resolved, but the shading leads to a decrease of the stars total brightness we measure on earth. The confirmation that it is indeed a planet, and not another, faint star in a binary system, was done using the Very Large Telescope.

The planet found by the students, a gas giant, is about five times as massive as Jupiter, the solar system’s most massive planet. It’s very close to it’s host star, with an distance of only 3 percent of earth’s. Even though by now there are over 300 known extrasolar planets, the discovery even is a special find: it’s the first planet discovered around a fast rotating, hot star.

Finding such an configuration is hard, as the planet’s orbit has to bring it exactly between the star and the observers (us). The most efficient way to find a planet is to look for shifting spectral lines. Stars are not fixed in space, but move similar to their planet. If the star and the planet were of equal mass, they would circle around each other (in fact, around their center of mass). Because stars are much more massive than planets, the center of mass lies almost at the same position as the star’s center. Thus the stars move only a bit. But this movement can be enough to shift the lines of a stellar spectrum according to the Doppler effect: when the stars moves into our direction, it appears bluer, when it moves away it appears redder. Think of the often cited emergency vehicle sirens: when the car drives into your direction the siren have a higher pitch, when it moves away a lower.
Now, as said the star moves only slightly, the shift of the spectral lines it thus very tiny. In a hot, fast rotating star, the spectral lines are broader (e.g. because one side of the star moves into your direction, the other away, so the lines will be broadened by a Doppler effect caused by the rotation). Thus it is hard to find a planet around a hot, fast rotating star with the classical method. But the new discovery shows that there can be planets orbiting around such stars.

Unfortunately the name of the planet has to follow the OGLE scheme and thus is OGLE2-TR-L9b. The three students would have preferred to name it after them: ReMeFra-1. Good to see they attached a number. Go on and find many more ReMeFras!

* For the OGLE-II database, 15700 stars were observed once or twice every night for four years. The acronym OGLE stands for Optical Gravitational Lens Experiment. It’s major goal is not to find planets, but to look for another acronym: MACHOs, massive compact halo objects. These are one possible explanation for dark matter. But the OGLE surveys have revealed several planets as a side effect.

You can find the corresponding paper on astro-ph: “OGLE2-TR-L9: An extrasolar planet transiting a fast-rotating F3 star”.

Tags: Astronomy, Discovery, Eclipse, Extrasolar Planets, Jupiter, Leiden University, Planets, Science, Star Gazer, Students, Study

Finally I’m finished :-):
In total I’ve listed 100 cases of Christian religious pareidolia, in which either Jesus Christ or the Virgin Mary have been seen. I will now analyse them in detail to see whether my vague theory mentioned earlier is supported by the data.

Long-Time developments and ratios

The rise and fall of Jesus’ Mother

Ok, I have to admit I do several things wrong in the figure. I shouldn’t combine the data points with a line graph and I should not include the 2008 numbers because the year isn’t over yet. And furthermore I should give error bars so one can judge the uncertainties. But let’s accept this for the moment.
First look at the overall percentages. Of the 100 events in the sample, 67 involve Jesus and 37 the Virgin Mary, in some cases both were seen. So Jesus is seen twice as often. Is this the case every year or are there trends in the data? Looking at the figure, one can see that there indeed is a trend visible: until this year Virgin Mary sightings increased and Jesus sightings decreased a bit. To make the numbers more comparable, let’s look at the rounded percentages of Jesus and Mary appearing each year are (they sum up to over 100% because sometimes both are seen at once):

Clearly sights of Virgin Mary had increased in the last years but the number drops strongly in 2008. As the year isn’t over yet, it is not sure if the trend will continue. But it seems as if we will have a Jesus dominated year again. I don’t have an explanation for this at hand, maybe it is just a statistical effect. (Please comment if you have one!)

Month dependency

Is Jesus a seasonal worker?

Another idea is that the frequency of sightings might be modulated depending on the time in the year. One would expect a maximum at the end of the year due to Christmas and maybe another at Easter, but the latter will be smoothed out because the date changes from year to year. The most current data from 2008 has to be excluded from the analysis because the year is not jet finished and further sightings will be reported in August, September, October, November and December. I might want do re-do the analysis in January next year.
For the years 2004 to 2007, in total 63 sightings are reported. I will not differentiate between Jesus and Mary because that would cause too few data points. When expecting a uniform distribution over the year and only statistical errors, the expected number of sightings is 5.25 per month with an statistical standard error of the square root of this. So within one sigma of about 2.3, 3 to 7.5 sightings are expected per month, quite a margin. Most month are well inside this, only July (0 sightings), August (12 sightings) and November (9 sightings) show deviations. I wouldn’t expect them to be very significant, the one sigma error only says that, expecting a normally distributed population, the probability to be within one sigma is approximately 68%. So 32% are outside. Out of 12 values we thus would expect about four to be outside, three are fine too. Furthermore the zero sightings in July will be made up with the nine in 2008 when including this year’s data. Nevertheless there seems to be a increase towards the end of the year, I’m curious how this will evolve in the next month.
The increase in the middle of the year, August, is quite significant (over 99 %). It might have to do with the media not having other things to talk about or maybe there is too much sun making peoples brains malfunction more often. I’m not aware of any as important Christian dates at that time than Christmas or Easter. But when one combines July, where no sightings are reported for the time span in question, and August, the numbers are within the expected average again. I think this is some kind of an artefact in the data.

Material and Texture dependence

Why can’t they simply eat Jesus?

The data do not show any strong trends for Jesus or the Virgin Mary to appear on different materials, with one exception: Only Jesus is seen in imaging like x-rays or ultrasound (7 times). In general, pareidolia perceptions are most frequent in food (29), then in wood (18), followed by dirt (11), light (10), on walls and in rock (8 each). Hearty, salty foods like potato chips or cheese toast are preferred.

The original theory was that Jesus is found more frequently in fluffy things and the Virgin Mary in flowing structures. This is well confirmed by the data as the following table shows. Appearances where the texture could not be determined because of bad pictures, no pictures at all or because I could not decide were excluded. These made up 14 pareidolia perceptions.

This clearly shows that Jesus is seen in fluffy structures three times as often as in flowing ones. The ratio is almost inverted for Virgin Mary, who is seen in over two out of three occasions in flowing textures. This backs my expectation that the usual pictures have something to do with the places where they are found. Virgin Mary usually having a head scarf or veil is found most often in flowing, fluid-like and soft textures that have less but bigger, more connected areas. Jesus in contrast, possibly owing to his beard and hair, is mostly seen in structures with rough surfaces, consisting of many different small areas.

Origin

Where does Jesus live today?

Investigating where the sightings occurred reveals that most are from the US. Only 16 are from other countries, four of those are from the UK (all Jesus, not surprisingly as there are many protestants), two from Australia and one each from Columbia, Germany, the Philippines, New Zeeland, Switzerland, Canada, Romania, Mexico, Ghana and Moldavia. There is a strong selection effect in the data as I only looked on English sites and news pages but I still think the US are especially ‘gifted’ concerning religious pareidolia.
Within the United States the distribution is also very strongly centred. Many states only have one or two sightings: Colorado, Indiana, North Carolina, New Mexico, Nebraska, Maine, Wisconsin, Columbia, Massachusetts, New Jersey, Kansan, Georgia, Missouri, Michigan, Idaho, Oklahoma, Tennessee, Connecticut and Virginia. I found tree each for Ohio, Arizona, Illinois and Pennsylvania. The top three states are far away from these values, they are:

• Place three: California with 13 total sightings
• Place two: Texas with 15 total sightings
• And the ‘Winner’: Florida with 16 sightings

Why are these three the top-states for pareidolia? Maybe it is because of an point mentioned earlier: They get to much sun in the south. Additionally, the number of inhabitants has to be taken into account. Simply compare the map above with this one.
Looking at the distribution of Jesus and Virgin Mary ‘appearances’ in these three states one can see a clear trend (double-counting sightings in which both are seen at once):

• California: Jesus 4, Virgin Mary 9
• Texas: Jesus 9, Virgin Mary 6
• Florida: Jesus 13, Virgin Mary 5

Texas shows a more or less average distribution, but California is clearly biased towards the Virgin Mary and Florida towards seeing Jesus. This is not unexpected as Julia from ForeignLight pointed out to me. In California there are more Catholics (the state has the largest Roman Catholic population of all states in the US) that have a connection to the Virgin Mary and in Florida there are more Protestants (40% of the population in contrast to 26% Catholics according to Wikipedia) that worship Jesus. Population in Texas is rather mixed, it has many Evangelical Protestants but also Roman Catholics. The numbers, according to Wikipedia, are: Protestants 32.5%, Catholics: 21% (Thus the ratio actually is exactly the Jesus to Virgin Mary ratio). I think it is an excellent and exciting result that these trends are also seen in the small data set analysed. It’s amazing to see the analogy.

What can we learn from this?

So how can I make money out of this crap?

As promised I now tell you where to look to find Jesus’ / Virgin Mary’s face. I think the most promising place is the classic one: Food (like a cheese toast). I’ll tell you why.

Often the pareidolia are found in food, especially baked, fried or toasted. I think one reason is that people have to eat frequently and while doing so they have the time to look. The Christians also have a special relation to bread (and fish, so Jesus seen in a fish stick is a classic).
Furthermore, in contrast to fixed objects like walls, you can change the perspective when looking at, say, a piece of toast. For Jesus or Mary to be seen in the structure of a wall they have to stand upright. The toast, in contrast, can be seen from different sides, you can rotate it and it has two different sides to look at! A toast also is a rather small area, so you are not distracted and the figure you want to see can be small. If you look at a wall the appearance has to be big so a passer by can see it. And most people eat more than one toast. Maybe there even is more toast out there than there are walls.
The light conditions outside change in the course of one day and from season to season, so if Jesus or Mary or any other face appear somewhere, that appearance might be gone an hour later. The light conditions while eating are simpler I think.
And the best: If you find a especially convincing depiction of a religious figure in, say, a slab of toast, it is easy to sell. You will have troubles to turn a public wall in an underpass into money and your might not want to sell your pet (which would not be possible via eBay either), so settle for the simple. Ok, it might not bee the most extravagant one, but referring to the Christians connection to bread it is convincing and it will be hard to come up with a reason why seeing Mary in a spot of mildew is flattering. And why should Jesus show up on only one toast?

I’d like to finish with pointing you to a more humorous approach of toast-appearances.

Tags: Cheese Toast, Cheesus, Christians, Faith, Food, God, Jesus, List, Perception, Religion, Sceptic, Special Events Weeks, Study, USA, Virgin Mary Cheese, Virgin Mary

UPDATE: I’m sorry, but some of the links are broken, probably because the news sites don’t keep their articles online forever.

Ok, as said before I collected pareidolia, false images our brains make up from patterns that have nothing to do with them. My idea is to analyse Christian reports of Jesus or the Virgin Mary seen on anything. My first result: These sightings are more frequent than I thought. It wasn’t that difficult to find these reports, actually I stopped after I found one hundred (100!) because I became bored and did not see any ending. The vast majority of them is from between 2003 and today. I list them in the table below and give you an analysis in my next post on the topic.

The ‘appearances’ of Jesus Christ or the Virgin Mary are sorted chronologically, the dates are given to the month, further detail I deem unnecessary. They are dated according to when the sightings were reported by the media, the original findings might be older. For example, the classic 2004 Virgin Mary Cheese Toast is said to have been reported ten years after it was made! As I can not check these claims I settle for the dates the articles were published. These might vary by a few days from source to source.

I collect these from the web, especially good sources were Bad Astronomy, the Angry Astronomer and StupidEvilBastard. Furthermore Wikipedia lists plenty of religious pareidolia. And then there is a collection on yoism. For Virgin Mary specific sightings, see VirginMaryAgain.

I try to include links to news articles or other sources as most of these ‘appearances’ are hard to believe and to meet your curiousness. It is hard to believe people are so ludicrous to see something in a dirty pan, reflected light, a potato chip or a t-shirt put to the laundry.
I sort the ‘sightings’ in categories.

• First, the region where they were found. Either the county, like Germany or the UK or, in case of the world record holder in religious Christian pareidolia, the United States of America, the state, be it Texas, Missouri or someplace else.
• Second, the material the object resembling (or, more correctly, resembling if you have a particular vision) Jesus or Mary is made of. That could be toast, ice, rock. Wood and bark are separated, I consider Wood to be cut so the figure is seen in the annual rings or branch knots and bark, on the outside of a tree, has a different texture and is more prone to light-and-shadow appearances.
• I also combine similar materials in categories. These are Food (e.g. toast, pancakes or seafood), Wood (anything associated with trees), Dirt and some more.
• Finally, I mention the general texture of the object or appearance and differentiate only between fluffy (not necessarily soft, but spotty) and flowing (soft, blurred or connected to fluid structures). Fluffy would be something like toast, rock or anything with a rough surface or made up of many small areas of different colour or brightness. Flowing in contrast is made by fluids (e.g. salt stains) or otherwise consists of larger, more connected or smooth areas, like those found sometimes wood and possibly dirt, depending on the actual appearance.

I will now list all religious perceptions I found in the quick search, including a small number of faked ones I like because there will always be someone who believes in them. This collection is by no means exhausting and probably biased towards the US because there are more media to mention them. I might add more when they are reported, but for the time being I have enough of them. I’ll present the analysis later next week, I hope.

Read the rest of this entry »