At Satapada the lagoon ends in a gap in a sandbar, and the gap is the whole argument. On one side is a brackish lake a thousand square kilometres wide. On the other is the Bay of Bengal. Between them is a channel a few hundred metres across that the sea keeps trying to close.

Everything about Chilika follows from that gap. Not the area, not the birds, not the dolphins: the gap. A lagoon is what you get when a river mouth is partly closed by the sea, and the degree of closure sets the salinity, and the salinity sets which species can live there. Close it further and you have a freshwater lake with a dead fishery. Open it and the sea comes back in, and with it the fish that spawn at sea and grow in the lagoon.

How it was made, and why it keeps changing

Chilika is young in geological terms — formed roughly three to four thousand years ago, when sea level stabilised and coastal sediment built a barrier across a drowned river mouth. Fifty-two rivers and streams drain into it from the Mahanadi delta and the western catchment, bringing fresh water and, crucially, silt.

The silt is the problem, and the mechanism is not complicated. Longshore drift moves sand northwards along this coast. The sand travels along the spit, reaches the inlet, and starts to close it. The outflow scours it open again. The mouth therefore wanders: it migrates along the shore, looking for the easiest exit, and the channel that connects it to the lagoon lengthens as it wanders. A longer channel means more friction, less tidal exchange, and a lagoon that becomes steadily fresher.

Plan of Chilika: the lagoon, the outer channel, the spit, the old mouth silting and the one cut in 2000 Bay of Bengal Chilika lagoon northern sector: under 50 cm deep sand spit, about 1.5 km wide 8 km of drift old mouth, silting up new mouth, cut Sept 2000 Schematic plan, not to scale. Channel 32 km before the cut.
The mouth is the meter on the lagoon's salinity. Longshore drift closes it, outflow scours it open, and the balance point keeps moving along the shore.

The 1990s, and the intervention

By the early 1990s the channel had lengthened and shoaled to the point where tidal exchange had dropped sharply. Salinity fell, fresh water weeds spread, and the lagoon began to behave like a lake. The fishery collapsed to somewhere around 1,600 tonnes. Ramsar had designated the site in 1981; in 1993 it was placed on the Montreux Record of sites whose ecological character had changed.

The response was unusually well prepared. The state created a development authority in 1992; the Central Water and Power Research Station at Pune ran hydraulic and two-dimensional model studies; the National Institute of Oceanography assessed the lagoon before and after. Their conclusion was geometric: the inlet had to be moved closer to the body of the lagoon, because a shorter channel means less head loss and more tidal flow.

On 23 September 2000 a new mouth was cut through the spit, 18 kilometres closer to the lagoon. The result was immediate and large. Fish landings went from about 1,600 tonnes to 4,889 tonnes in 2000–01, and to 11,989 tonnes in 2001–02. Salinity gradients came back, seagrass returned, the dolphins had their estuary again, and in 2002 Chilika became the first site in Asia to be taken off the Montreux Record. The authority received the Ramsar Wetland Conservation Award the same year.

And there are people behind the tonnage. Between 150,000 and 200,000 fisherfolk live in and around the lagoon, with several hundred thousand more earning from it indirectly — prawn, crab, weed, boats. The development authority's own account is that average household income among fishers rose by more than fifty thousand rupees a year after the new mouth opened. Nalabana island, 15.57 square kilometres in the middle of the lagoon, is a notified bird sanctuary that hosts migrants by the million in winter. None of that is an abstract ecological indicator. It is income, settled by season.

What the success story leaves out

Nothing in that account is wrong. What it leaves out is that the forces which closed the old mouth did not stop when the new one was opened. Longshore drift kept carrying sand north. The new inlet started migrating, as inlets do, and by the 2020s it had moved roughly eight kilometres north along the spit and narrowed. Sediment kept arriving — on the order of 800,000 tonnes a year, according to the state forest minister's reply in the assembly — and the internal channels kept filling.

So the salinity began falling again, the weeds began returning, and the government is now preparing another intervention: a detailed project report, costed at around two thousand crore rupees, for a new mouth, desilting of the critical channels and a monitoring regime with maintenance dredging every two years.

This is not a failure of the 2000 project. It is what the 2000 project was: a reset of a balance that has no resting point.

Restored, or maintained?

The vocabulary is the interesting part. Restoration implies a prior state and a finish line — you restore a building, and then it is restored. A tidal inlet has no finish line, because its geometry is the output of two processes that never stop: sediment coming in, water going out. The equilibrium is dynamic, which is a polite way of saying it is always somewhere else.

Which means the honest unit of management is not a project but a schedule. A lagoon like this needs a survey every few years, a dredge on call, and someone with the authority to cut a new mouth when the old one stops working. That is a maintenance contract, and maintenance contracts are politically unattractive because there is no opening ceremony for the fourth desilting.

There is a second thing the story leaves out, and it cuts the other way. The mouth has always wandered; the lagoon has been fresher and saltier many times before anyone was counting fish. What is new is not the change. It is the dependence — a fishery, a dolphin population, a tourist economy and a few hundred thousand people whose income is calibrated to a particular salinity in a particular decade. The lagoon did not become fragile. It became specified.

That is the trap in every coastal restoration. You cannot restore a system to a state and also keep the yield that state produced, because the yield depended on holding the system still — and the system is made of water that will not hold still. What you can do is decide which state you are buying, and budget for the fact that you will have to buy it again.

Sources and checks

  1. Chilika Development Authority Eco-restoration of Chilika Lagoon: hydrological intervention, monitoring and outcomes
  2. Chilika Development Authority / ICAR-CIFRI Fish and shellfish diversity and its sustainable management in Chilika Lake
  3. Ramsar Convention Chilika Lake: designation 1981, Montreux Record listing 1993 and removal 2002
  4. Central Water and Power Research Station, Pune Hydraulic and two-dimensional mathematical model studies of the Chilika inlet
  5. National Institute of Oceanography, Goa Pre- and post-intervention environmental impact assessment of the new mouth
  6. Odisha Forest and Environment Department / IIT Madras Preliminary project report on siltation, inlet migration and a proposed new mouth, 2025

Last checked 2 November 2026 Yellowsea Huang Corrections