Seismic exploration for pharaonic alabaster


There’s a bit of confusion around the name alabaster. Strictly speaking, for geologists it indicates a fine-grained gypsum. But archaeologists call alabaster the fine-grained, banded deposits of calcite used by the ancient Egyptians for statues, jars, canopic vases. The ancient Egyptians carved this alabaster stone to make small ointment jars called alabastri, in a town called Alabastron: the question ‘what gave the name to what’ is more complicated than a chicken-and-egg problem.

Anyway, the Egyptian alabaster is actually a whitish translucent travertine: calcium carbonate cave deposits concretions, dripstones and flowstones. When cut in thin slices, it reveals its many colours (yellow, orange, white, tan, green and red) and shows its translucent nature. In modern craft it is used for lamps: like the three in the title photo.

It was quarried from the early Dynastic period in the Nile valley to build pavements, statues, jars, like the incredible pieces found in the Tutankhamun tomb ( The old pharaonic quarries are known and mapped; this post is not about a high-resolution seismic survey to find more alabaster down there. It is more about the shape of the bands of this stone, about the patterns often found in seismic sections in thrust-belts, and about the bandwidth of seismic data. Look at this seismic section.


Is it realistic enough? A structural geologist might spot the trick. In fact the layers are not folded, but deposited with this shape, drop by drop. This image is actually a synthetic seismic section generated from a photo of an alabaster slice: we use the average RGB intensity as acoustic impedance, estimate the reflectivity, convolve with a wavelet, and display in grayscale.

The resolution is low! It’s low-frequency! someone will complain. Yep, true. I used a simple seismic wavelet with a rather low central frequency. If we repeat the exercise with a shorter wavelet, we get an image with more detail: and we resolve, for example,  boundaries that are close to each other.


I am aware that all this is pretty basic, but I would like to show how a very broadband section looks like, and the fine detail that can appears on a grayscale broadband ‘seismic’ image. The very low frequency component gives the aspect of a three-dimensional carving: the low frequency bright and dark zones seem shadows on a bas-relief. When you have enough low frequency, different layers appear differently, and you see more than just the boundaries between them.


The original starting photo from which all started (in colour)  is the one here below. There is light behind the alabaster slice.


The fake seismic section is more than the photo in black and white, since its generation involves derivatives: it enhances details and depth.

The alabaster is polished: but if you overlay the broadband section and the photo, you get the impression of rough layers on an outcropping cliff.



The biggest problem of these images are the greasy finger marks that people will leave on your screen: they can’t keep from touching the image to feel if it’s rough, to feel if they feel the texture.







Leave a Reply

Your email address will not be published. Required fields are marked *