We have been working on this project for a while now and finished it just recently. Prof. Butterlin from the University of Paris commissioned us to recreate an excavation trench of Mari in 3D. The site Tell Hariri, situated in modern Syria, is unreachable at the moment and the excavation team needed a tool to work with the stratigraphy of the most important trench V.1.
We worked closely with Wael Abu-Azizeh, the leading excavator of that trench, to rebuild the stratigraphy and architectural remains in the computer. The results were final renderings and… a Google SketchUp model! You can read everything about the project here.
Interesting; this is what all our documentation is supposed to provide us, but too often we end up with documentation that is only 2D (the profiles visibile in the photo), rather than 3D of surfaces, and perhaps some kind of indication of what is in the layers (volumes rather than just surfaces), which might provide evidence for things like post-depositional transformations.
One thing I’ve always had trouble conceiving was how to work with unclear layer boundaries in 3D CAD, for example.
The simplified forms make rendering easy, and probably aid with conceptualisation.
But I assume you can’t strip off each layer, one by one, and “re-excavate” it, only visualise the finished product?
Well, it is all dependent on the detail of documentation of course. But under normal circumstances, a layer-by-layer visualisation is not possible, because the amount of data that one has to collect for that would be too immense.
What was possible in Mari though, was the detailed visualisation of different chronological periods and even architectural sub-phases. We should not forget, that in the case of Mari, it was not intended to rebuild the trench in 3D at first. It was a necessity in order to be able to work on the stratigraphy without going back to the trench in Syria.
Only this way, inconsistencies were visible and foremost understandable. I think this is the strength of that kind of visualisation.
“Only this way, inconsistencies were visible and foremost understandable.” Often a problem with legacy data, and conversion from one format to another.
Also, as has been noted for over 100 years, you’re always missing something, and rarely notice till you come to doing post-ex.