The submerged cities discovered by India’s National Institute of Ocean Technology in the Gulf of Cambay are around 40 metres deep.  Basic knowledge of earth sciences leads to the conclusion that, by virtue of their great depth, the Cambay cities are very probably more than 6,000 years old – and hence stand in contradiction to orthodox views on the origins of civilisation.

But as part of our research for Underworld, we wanted to know whether it would be possible to get more precise lower limits on the dates of submerged ruins.  For this we went to geologist Dr. Glenn Milne at the University of Durham, who has developed a state-of-the-art model of sea-level change.  Glenn’s model takes into account a number of major processes that affect sea-level and it can date the submergence of underwater ruins anywhere in the world with relatively good precision and reliability.  Generally speaking, it can estimate the date of submergence to within 1000 years.  An important exception concerns areas that have been subject to large-scale vertical tectonic motion.  Such motion cannot be accurately modeled at the global level.  Since Cambay is a tectonically active area, we have to keep tectonics in mind when using Glenn Milne’s sea-level model to estimate the minimum age of the Cambay cities.

The first approach we took involved inundation maps.  Glenn Milne made these maps by applying his sea-level model to topographic (or bathymetric) data for the Gulf of Cambay.  By knowing how deep the gulf is today, and by also knowing how this depth has changed since the end of the last Ice Age, Glenn Milne can produce graphical maps showing what land was exposed at various points in time.  These maps suggested that the Gulf of Cambay was only formed between 7,700 years ago and 6,900 years ago.  The map for 7,700 years ago showed the gulf to be dry land, whereas the map for 6,900 years ago showed it to have been flooded by the Indian Ocean, creating the gulf that we recognize today.  In Underworld and our subsequent correspondence with the press we have put forward the date of 7,000 years as a very conservative minimum amount of time since submergence of the Cambay cities.

In recent weeks, Glenn Milne took an alternative approach to putting a minimum date on the submergence of the Cambay cities.  Instead of using inundation maps, Glenn Milne ran his model on a specific pin-point location.  Inputting the approximate long/lat co-ordinates of the cities, followed by the NIOT’s empirical measurement of the depth of those cities (40 metres) produced a time since submergence of 12,000 years.  This result is more compatible with what one would expect given knowledge of the depth of the site and of late glacial sea-level rise.  However, there is clearly a huge of difference between this estimate and the earlier one of 7,000 years.

How did this disparity come about?

And why did we go with the date of 7,000 years in Underworld even though it is, arguably, excessively conservative?

The answer to the first question is simple but unexpected.  To create the inundation maps, Glenn Milne applied his model to a global electronic topographic dataset.  This dataset is called Terrain-Base.  TerrainBase is state-of-the-art and used in top universities around the world, but since our planet’s oceans have not been fully explored and bathymetric data about them has not been fully compiled and converted into an electronic format, this does not make it totally reliable.  (The dataset is public and can be accessed through America’s National Geographic Data Center  We did not expect the topography to be wildly inaccurate, but we now know that it is – at least for the region we’re interested in here.  TerrainBase tends to underestimate the depth of the Gulf of Cambay by 15 to 20 metres relative to a 1993 sounding chart made by the US Defense Agency.  So as far as Glenn Milne’s model was concerned when making the inundation maps, the Gulf of Cambay was much shallower than it actually is and was submerged much later than it probably was.

There are a number of reasons why we went with the conservative date of 7,000 years for the submergence of the cities.  Firstly, towards the beginning of our interest in the Gulf of Cambay, we were not absolutely sure whether the depth of 40metres given in the Indian press reports was accurate or whether perhaps it was a maximum depth reached by specific parts of the cities – parts that may have been too small to show up on the inundation maps.  At that time we were more inclined to trust the TerrainBase bathymetry, and had no real reason not to.  Secondly, even as involvement with the Cambay story progressed, our interested in it remained primarily archaeological rather than geological.  The cities would still be pre-Harappan and anomalous (relative to conventional views on the origins of civilization) regardless of whether they were submerged 7,000, 10,000 or 12,000 years ago.  We therefore continued putting forward the date of 7,000 years while explicitly emphasizing that it was a conservative minimum period of time since submergence.  Thirdly, in the absence of direct information on tectonics in the Gulf of Cambay, we felt that the date of 7,000 years was perhaps not excessively conservative.  What if some huge (but hypothetical) earthquake had caused the cities to subside by 15 metres at some time during the last 7,000 years?  From the point of view of Milne’s model, this would mean that the cities were only 25 metres deep rather than 40 metres, because Milne’s model does not take tectonic motion into consideration.  Therefore and given our specific archaeological interests, the possibility of extreme tectonic subsidence at least partially cancelled out TerrainBase’s 15-20metre underestimation of depth in the Gulf of Cambay.

Given the location of the cities, it is very probable that they have been affected by vertical tectonic motion to some extent.  However, there is no direct evidence to show that a large proportion of the depth of these cities has resulted from tectonic subsidence.

The fact that the cities are still in tact strongly suggests that tectonic motion has been neither massive nor abrupt.  Certainly, anything more than ten metres of subsidence in the last 7,000 years would be highly unlikely and without geological precedent.   Ten metres of tectonic subsidence would still leave 30 metres of depth to be accounted for by slower eustatic and isostatic processes, such that a late glacial date for submergence of the cities would still be most probable.   (And after all, the most likely time for dramatic tectonic activity would not be in the last 7,000 years, but around the late glacial period when sea-levels were in their greatest state of flux and causing greatest disruption to the forces on the earth’s crust.)

If tectonic subsidence has played any role in the subsidence of the cities, then Glenn Milne’s later result of 12,000 years would provide an upper limit on the time since they were submerged.  Combined with our conservative lower limit of 7,000 years, we would expect the cities to have been submerged at some point between 10,000 and 5,000BC.  Indeed, the carbon-dates released recently by the NIOT fit well within these limits.  The NIOT’s carbon-dating placed a piece of cut wood, recovered from a shallow layer on top of the city, to 7,500BC (calibrated). 

Does this mean that the cities themselves are 9,500 years old?  Yes – they are very unlikely to significantly younger than that.  But, for a number of reasons, we suspect that the cities might turn out to be even older.  Firstly, the cities are enormous and sophisticated, so they could not have sprung out of nowhere.  This means that there could well be layers of habitation deeper and older than that which yielded the cut piece of wood that was carbon-dated.  Secondly, though we have no direct evidence on tectonic motion in the precise location of the cities (and will not have until core-sampling has been carried out by the NIOT on a future expedition), we are inclined to suspect that such motion has contributed a lot less than 10 metres to the depth of the cities, meaning that submergence may have happened closer to Glenn Milne’s upper limit of 12,000 years ago than to his lower limit of 7,000 years ago.  Part of this suspicion arises from an article by Prasad et al. (Geomorphology, 1998, vol.25, no.3-4, pp.207-223) which suggests that the land in the region of Lothal (near the Gulf of Cambay) has actually experienced tectonic uplift since the incursion of the gulf.  (See this post by Dr. Ward Sanford on our MB: ) Lastly, carbon-dates for artifacts that have been submerged in salt-water for a long period of time may possibly underestimate the age of those artifacts.  This means that the piece of wood might possibility be even older than 9,500 years.