The Pyramids and Orion's Belt
I've come across many claims that the three great pyramids at Giza (Pyramids of Khufu, Khafre and Menkare) mirror the pattern of stars in Orion's Belt. This is fairly evident, just from looking at them. They are in almost a straight line, bent at a slight angle. However, various people seem to take this to an extreme, saying that the alignment is very accurate. Others read even deeper into this, saying the alignment isn't exact now, but was much closer when the pyramids were built (around 2550 BC), and would have been perfect in around 10,500 BC. This leads some to conclude that the pyramids were planned way back in 10,500 BC, and gets into the theory of how the Egyptian Civilization is much older than modern scholars believe.
I decided to investigate this theory, with some basic tools and the Internet. The image below is an overlay of two images. The base image, showing the stars in Orion's belt was produced by my planetarium program (Starry Night Pro Plus v.6.2.3), set to the latitude of the Giza Plateau, 2008 AD. The overlay image is an Iconos satellite photo of the Giza Plateau. Green circles indicate the centers of the 3 pyramids (click the image for a larger version). Note that the green circles are skewed off the tips of the pyramids because the satellite image was taken at an angle, not when the satellite was directly overhead.
As you can see, the 3 stars form an almost horizontal line, which is slightly bent upwards. However, the line formed by the centers of the 3 pyramids is not so flat - it forms a more acute angle than the line formed by joining the 3 stars. So while the pattern formed by the pyramids resembles the pattern of stars in Orion's belt, it is by no means very accurate.
What about the claim that the correspondence between the stars and the pyramids was much closer in 10,500 BC? I dialed back Starry Night to show the sky from Giza in 10,500 BC, to compare it to what it looks like today. Stars are not stationary with respect to the earth, in fact they are constantly in motion. As a result of this motion, their pattern in relation to each other changes over time, as viewed from the earth. These changes are very small, and take a long time to become apparent, because of the enormous distances.
The picture below is an overlay of three images produced by Starry Night. The images are of the night sky as seen from Giza on 3 different dates: 2560 BC, 10500 BC, and 17000 BC. All are at the same magnification (snipped from originals which showed a 3 degree x 2 degree view of the sky). In order to distinguish between them, I colored the stars in the 2560 BC image white, the stars in the 10500 BC image yellow, and the stars in the 17000 BC image blue. I then overlaid the images one on top of the other (applying an appropriate transparency to the images on top) and registered them to each other using the following procedure: first I overlaid the central star in the belt (Alnilam) directly on top of the same star in all 3 images, and then I rotated the top two images individually to align the leftmost star (Alnitak) in a direction as close to the bottom image (2560 BC) as possible. I wanted to see how the angle of the line formed by joining the 3 stars changed, at the 3 different dates.
Again, you can click on the image above to see a higher resolution image. I have also zoomed in on the 3 stars individually in the images below:
As you can see, there are a couple of interesting things to note:
First, the apparent lateral motion of these stars over time is very small. Initially, I only had two images: from 2560 BC (white) and 10500 BC (yellow) to test the theory. The change in relative positions of the stars over this time period was so small, that I could not believe it would be anywhere near sufficient to provide an exact match to the position of the pyramids. As is evident in the first image on this page (the pyramid-star overlay), a much more drastic change of position is required in the stars to match the exact position of the pyramids. For this reason, I added a 3rd image, going back in further in time to 17000 BC (blue). Even then, the stars do not match the position of the pyramids exactly.
Second, and more interesting, consider the direction of change. It's clear from looking at the overlays that the line joining the 3 stars is becoming flatter, the longer you go into the past. This is exactly the opposite of the change that is needed to align the stars with the pyramids. You can see in the first image that the line joining the pyramids subtends a more acute angle at the center. In fact, the line joining the stars needs to acquire a more acute angle to match this. However, when we go back in time, exactly the opposite happens. The line joining the stars broadens instead. Further, the deeper you go back in time, the flatter the line gets, as can be seen by comparing the 10500 BC positions with the 17000 BC positions.
I don't know what is going on here. Either these people are wrong, or Starry Night is wrong. Or I've made a mistake in my methodology. If anyone sees any flaws, let me know.