There is a compound wall in Chembur that used to be a mudflat, grey-brown mud cracked into plates where the tide reached and withdrew twice a day, and my friend remembers it that way still, though nobody younger than her does. She believes Mumbai’s mangroves are being quietly murdered for real estate. She has no data for this, only the accumulated evidence of her own eyes : the wall where the mudflat was, a creek nearby that used to smell of salt and now smells of construction dust. She asked me to look into it, mostly, I think, to have the number to be angry with.
So I went looking for the number, and it turns out there is some sleuthing involved. Satellite imagery and rigorous mangrove surveys are a recent phenomenon; before that, you are reconstructing a coastline from whatever the British happened to draw. The reconstruction is an estimate rather than a census but a reasonable figure can be built: Mumbai’s mangroves covered something in the order of 180 square kilometres at the turn of the twentieth century. By 1990, harmonized satellite baselines put the figure at 92 square kms. A hundred years, and the city had already lost roughly half of what it started with.
In itself, this is damning, a clear indictment of short-term developmental thinking over ecology, exactly what my friend feared. But that would be a conclusion drawn too quickly. If all one knew was the century-long collapse, the story would seem obvious, a straight line from empire to independence to real estate, drawn by the same hand the whole way. History rarely arranges itself neatly.
From 1990, the record gets more granular and does something it doesn’t prepare you for: the rate of erosion keeps falling, slower each time, almost politely. Ninety-two square kilometres in 1990. Seventy-six by 2000. Seventy-one by 2010. Seventy by 2015, essentially unchanged five years later. Sixty-nine in 2026. Something interrupted the trajectory, and the interruption itself has a name: Coastal Regulation Zone enforcement, tightened through the 2000s, and a run of court interventions that did not let the matter drop the way such matters usually do. So the cause, unlike in most environmental stories, isn’t really the mystery. What’s uncertain is something else: whether an enforcement regime that has held for three decades is being held up by something durable, or whether it has simply not yet met the year, the project, the government, willing to test it.
Before the arithmetic gets any more comfortable with itself, it’s worth saying what the sixty-nine square kilometres are actually being asked to do. The sea off the Konkan coast has been rising through every one of the years the mangrove line held flat. The monsoon arrives now in fewer, heavier days rather than a generous, distributed season. A wall of roots that hasn’t shrunk is still a smaller wall than it was, measured against a sea that hasn’t stopped rising and a sky that has stopped being patient about how it delivers its water. The 2026 figure has to do more work than the 1990 figure did, and nobody has adjusted the number to reflect that.
Call it stable, if the word survives contact with the arithmetic. A quarter of what remained in 1990 is gone by 2026; twenty-three of ninety-two square kilometres, lost not to a single decade of reckless growth but to thirty-six years of the city quietly managing to lose a little, again and again, under a regime everyone agrees is working. The 2026 figure is the tell. Sixty-nine, down one from 2015, described in the record not as decline but as net stability despite project-specific clearances is a genteel way of saying the aggregate held its shape while individual pieces of it were still, quietly, given away.
A policy is a sentence a government writes about itself. By that yardstick for mangroves then it is a suspended sentence. The line could hold at sixty-nine for another decade, or it could not, not because the regulation itself would need to be rewritten, but because someone with the standing to enforce it would only need to decide, one budget cycle, not to bother.
Widen the frame further and the pattern doesn’t disappear, it changes sign. Mumbai’s mangroves sit at one end of a coastline that runs seven hundred and twenty kilometres from Palghar down through Thane, Raigad, Ratnagiri, Sindhudurg — six districts, the tidal range shrinking steadily as you move south, the same species of forest metabolising a slightly different sea in each one. Look at the state’s own survey, taken across fourteen years : Maharashtra’s mangrove cover grew sixteen percent. Raigad alone added something in the order of twenty-two square kilometres. Mumbai and its suburbs, twelve point nine.
Read past the headline number and Palghar tells you what the number is hiding. The district added eight square kilometres of mangrove in the survey period and lost nineteen of a category called sparse mangroves, the thin, young, easily bulldozed fringe that upgrades into forest proper if left alone, or vanishes if it isn’t. A forest losing its sparse edge is a forest whose tide has stopped reaching as far inland as it used to. The state’s own Mangrove Cell, asked to explain a number that flattered it, allowed that some of the apparent gain was mangrove colonising mudflat, narrowing the creek behind it — Thane Creek especially — the same mechanism, run in reverse through the accounting, that shrank Malad Creek from thirteen square kilometres to under ten.
So the wider frame doesn’t rescue the number, it explains it. A city can describe a slowing loss as stability. A state, it turns out, can describe the identical throttling of a tideline as growth, and both descriptions will be true of the arithmetic and false of the water. Nobody is lying in either case. They are only choosing, with each shift in scale, which noun gets to carry the sentence.
But every comparison so far, city against state, decline dressed as stability against growth dressed over decline, has conceded something it should never have granted: that a square kilometre of mangrove is fungible with any other square kilometre of mangrove, the way a bank balance survives a bad month if enough good ones follow it. Ecology keeps no such ledger. A clearance that removes less than one percent of the city’s remaining stock by area can still be the loss that matters most, if what stood there was the one tidal channel juvenile fish used to cross into the bay, or the fringe that broke a surge before it reached a settlement with no other wall between it and the sea. The percentage was never measuring the thing that mattered. It measured area, and reported it as though area were a proxy for function, the way counting hospital beds is not the same as counting doctors.
And this loss does not decelerate the way the citywide number does, and it is not restored by the accounting trick that flatters the state’s total. A dense mangrove forty kilometres north does not grow gills for the juveniles that lost their channel. A mudflat that becomes a compound wall does not become a mudflat again because a different creek is being colonised by mangrove advancing over what used to be open water — the very process the Mangrove Cell counted as gain. The state’s number can rise. The nursery is still gone, and nothing in the ledger notices the difference.
This is the sentence within the suspended sentence, the part the arithmetic cannot reach: some losses are never on the ledger at all, because no one measured what a particular hectare was doing before it was cleared, and by the time anyone thinks to ask, the answer is a compound wall. She wanted a villain. What she has instead is a suspended sentence, and one clause inside it that nobody transcribed: her creek, the one that used to smell of salt. It isn’t in the sixty-nine. It never was. It just smells of construction dust, and will keep smelling that way whatever the number says next year.