Coast & Islands
India’s ancient ports moved because rivers silt up and coastlines shift
The great trading harbours of the past sat on river mouths, and the same sediment that made those mouths navigable eventually closed them, which is why several famous ports are now inland or gone.
By Pranav Kulkarni3 min read
A river mouth is the obvious place to put a port and the worst place to keep one
Before deep-water engineering, a harbour needed shelter, depth and a route inland, and a river estuary supplied all three at once. So the old ports of the subcontinent sit on river mouths almost without exception, from Gujarat round to Bengal.
The problem is that a river is also a conveyor of sediment, and it deposits that sediment precisely where its current slows — at the mouth. Left alone, a working estuary silts steadily, the channel shallows, and the depth that made the port possible gradually disappears.
Add to that the ordinary mobility of a delta, where the main channel can shift to a new course entirely in a flood, and the surprise is not that ancient ports were abandoned. The surprise is that any of them lasted as long as they did.
Muziris is the famous case and it is not fully resolved
Classical sources describe a major pepper port on the Malabar coast, known in Greek and Roman writing as Muziris, trading with the Mediterranean world. It appears in Tamil literature as well, and then it stops appearing, and its location was uncertain for a very long time.
Excavations at Pattanam in Kerala have produced material consistent with a substantial ancient trading settlement, and it is widely discussed as the most likely site, though the identification is debated among archaeologists rather than settled.
A flood of the Periyar in the fourteenth century is often cited as having reshaped the coast and finished the port, and the same event is credited with opening the harbour at Kochi that replaced it. This account is traditional as much as documentary, and how much of the change was one event and how much was gradual is exactly the kind of question that is hard to answer from silt.
The northwest lost several harbours the same way
The Gulf of Khambhat coast in Gujarat held some of the most important ports of the medieval and early modern Indian Ocean. Their decline is generally attributed to a combination of silting in a very high-tidal-range gulf, the shifting of river channels and the arrival of ships that drew more water than the approaches could take.
Lothal, much older again, is a Harappan site with a large brick basin that has been interpreted as a dock connected to a river channel. The interpretation is contested — some archaeologists read the structure as a water tank rather than a dock — and the debate has run for decades without resolution.
Either way the geography makes the point. Lothal is not on the sea now. Whatever the basin was, the water that reached it does not reach it any longer, and the coastline that mattered to the people who built it is somewhere else.
Modern ports solved the problem with machinery
The reason today’s major ports can stay where they are is dredging. Channels are cut and maintained continuously, breakwaters redirect the movement of sand along the shore, and depth becomes an operating cost rather than a natural gift.
That intervention has consequences down the coast, because a structure that traps sand on one side starves the shore on the other. Erosion downdrift of harbour works is a recurring and well-documented pattern on the Indian coastline, and it is one of the reasons some fishing villages have lost beach in living memory.
So the ports have not escaped the sediment problem. They have converted it into an engineering and political problem, which is a different thing from having solved it.
Visiting the old ports is mostly an exercise in imagination
The practical warning for a traveller is that these places rarely look like anything. An ancient port site is generally a low mound, an excavation trench, a small site museum and a landscape that no longer contains the water it was built for. Going without context is a reliable way to be underwhelmed.
What repays the visit is the local museum and the surrounding geography rather than the site itself. Standing at an old port and working out where the sea used to be, and how far away it is now, is the whole point of the trip.
Site status, access arrangements and whether excavation is open to visitors all change, so the current position with the relevant archaeology department or state tourism office is worth confirming before travelling out. Some of these sites are extensively presented. Others are a signboard in a field.
Common questions
Is Pattanam definitely Muziris?
No. It is the leading candidate and the excavated material is consistent with a large ancient trading site, but the identification is argued over by archaeologists and has not been established beyond dispute. Anyone presenting it as certain is going further than the evidence currently supports.
Why did the coastline change so much?
Rivers deposit sediment at their mouths, deltas shift their channels, storms and floods rearrange the shore, and relative sea level has not been constant. Over the centuries that ancient ports operated, all of these acted together, and the cumulative effect can move a shoreline by kilometres.
Are there old port sites worth visiting for a general traveller?
A few, particularly where there is a good site museum or where the surviving town has visible history of its own. As a rule, treat them as part of a wider trip through the region rather than as destinations in themselves, since the archaeology alone is often thin on the ground.
Editor, Traveology
Pranav writes the explanatory pieces on states, hills, coast & islands and reads the small print so you do not have to.
