Dholera environment plan: waste, effluent and the 2014 clearance

The parts of a new city that decide whether it survives are almost never the parts anyone photographs. Nobody puts a common effluent treatment plant on a brochure cover, and no sales presentation has ever opened on a bio methanation unit. In the time I have spent reading Dholera material, official and promotional, I have seen the fab rendered from every angle, the expressway shot from a drone, the terminal building filmed in soft evening light, and the waste system mentioned roughly once per document as a line of capacities that nobody then expands on. That silence is strange, because if you read the record of cities built from scratch anywhere in the world since 1980, the environmental layer is where more of them ran into serious trouble than any brochure would lead you to expect.

So this is the essay about drains, effluent, garbage and a clearance certificate dated 2014. I think it is one of the more useful pieces on this site, for three reasons. The published inventory is unusually specific, which tells you something about the people who sized it. That inventory is a starter set for roughly 22.5 sq km rather than a system for a million residents, which almost nobody says out loud. And the single most consequential item in the whole environmental layer, the clearance itself, is a document I can date but cannot read the current version of, which is exactly the sort of gap this site exists to name rather than paper over.

Three waste streams, not one problem

People talk about waste as though it were one thing. In an industrial city it is at least three things, and they are unalike in engineering, in cost and in who carries the legal risk.

Domestic sewage is what households and offices produce. It is biologically predictable, it arrives in daily rhythms, and treating it is a mature, well understood problem. Industrial effluent is a different animal entirely, because its composition depends on what the factories upstream are making, and a plant that treats one recipe well can be defeated by another. Solid waste is a logistics problem before it is a treatment problem: it has to be collected, separated, moved and then processed or buried, and every stage depends on the discipline of the stage before it. Sitting above all three is a fourth layer that is not infrastructure at all, the regulatory one: the clearance, the conditions attached to it, and whoever is supposed to be monitoring compliance against them.

Dholera has a published answer for each of the first three. On the wastewater side, the sanctioned material describes a 10 MLD sewage treatment plant and a 20 MLD common effluent treatment plant, with roughly 81 km of recycled water pipeline to carry treated output back into use. On the solid waste side it describes a chain: 25 TPD of segregation, 30 TPD of bio methanation, two 25 TPD incinerators, and a 28 ha integrated landfill. Those figures come from the DSIRDA sanctioned development plan and a 2024 DICDL interview, so I carry them as plan and reported tier rather than as audited operating results. The one hard primary marker sits underneath them: the NICDC Delivery Monitoring Unit report to DPIIT dated 30 June 2026 records trunk infrastructure works as complete in the Activation Area, the roughly 22.5 sq km plug and play starter zone inside TP2. The same report records the project's Environmental Clearance as dated 19 September 2014.

The plant that matters most is the one nobody has heard of

If I had to pick one item from that inventory as the most consequential for whether Dholera works as an industrial city, it would be the 20 MLD common effluent treatment plant, and I doubt one buyer in a hundred could tell you what it is.

A common effluent treatment plant is shared industrial plumbing. Individual units pre treat their own discharge and then send it into a central facility that does the heavy treatment for everybody. The reason this model exists is economic: a small or medium manufacturer cannot realistically build, staff and run a compliant treatment plant of its own, so a cluster pools the cost. When it works, it is one of the better ideas in industrial planning, because it makes compliance affordable for firms that would otherwise quietly not comply.

The catch is that it converts a private problem into a shared one. Every unit connected to that plant depends on it running properly. If the plant is overloaded, or if one connected firm sends in something the process cannot handle, the failure is not contained to the offender. It lands on the whole cluster at once, and on the operator of the plant. For an occupier deciding where to put a factory, the health of the shared effluent system is not a footnote, it is a live operating risk that sits outside their own gate and outside their own control.

What makes it interesting at Dholera specifically is the breadth of what the plan intends to attract. The official sector list runs to Defence, Aviation, Electronics and Semiconductors, High Tech, Pharma and Biotech, Heavy Engineering, Auto and Ancillary, General Manufacturing, Agro and Food, Metals, and renewables. The plan's own sector job potential figures put Electronics at 87,300, Pharma at 49,100, Heavy Engineering at 45,100, Auto at 43,900, General Manufacturing at 42,400, Agro and Food at 27,500 and IT at 6,200. Read those as an effluent problem rather than an employment one and you get a difficult mix. Pharmaceutical effluent, metals effluent, food processing effluent and semiconductor rinse streams do not share a treatment recipe. They differ in what they carry, in how concentrated it is, and in what neutralises it.

None of that means the design is wrong. It means the design is demanding, and that the acceptance rules governing what a connected unit may discharge into the shared plant will matter more here than they do in a single sector industrial estate. What I cannot tell you, because I have not found it in any document I would put my name to, is the discharge standard the plant is built to, the route treated effluent takes after it, or whether zero liquid discharge is intended for any part of the zone. That is not reliably sourced, and I would treat a confident verbal answer on it as a claim to verify with the planning authority rather than a fact to repeat.

Twenty MLD is also a starter number, and should be read as one. The plan describes a total industrial area of 11,000 ha with roughly 3,000 ha of it in Phase 1, and industrial parcels ranging from 0.5 to 150 hectares. Twenty MLD is sized against the zone where trunk works are recorded complete, not against 11,000 hectares of eventual factories. I am not going to divide one by the other and pronounce a verdict, because staged capacity for later phases is not something I can source. Directionally, though, the honest statement is that the effluent system that exists is a first module of something much larger that has not been built or, as far as I can see, publicly costed.

Load is beginning to arrive, which is what makes this worth watching now rather than in 2032. The DMU report records 14 plots and 545 acres allotted, of which 476 acres are industrial, with Tata Chemicals named as an anchor industrial allottee, and a further 1,043 acres of industrial land plus 1,031 acres of other land ready for allotment. I am neutral on every company involved. The point is structural: the first real effluent load in this city's history is on its way to a plant that has, until now, had very little to do.

Sewage, and a plant waiting for people

The 10 MLD sewage treatment plant is the mirror image of that problem. It has capacity and almost no customers.

The residential arithmetic is brutal and worth stating plainly. Census 2011 recorded 2,779 people in Dholera village across 576 households. The earlier target of roughly 120,000 residents and 80,000 jobs by 2020 lapsed unmet, and no resident wave has arrived since. A sewage plant is not a warehouse; it is not indifferent to sitting empty. Biological treatment stages depend on a reasonably steady organic load to sustain the microbial population that does the actual work, which is why commissioning a plant and ramping it up is a real operational exercise rather than a switch. I do not know which process this particular plant uses, and the published material does not say, so I will not push that observation further than it deserves. The general shape of it stands: a treatment plant built for a population that has not arrived is an asset in waiting, not an asset in service.

That 10 MLD against 20 MLD for industrial effluent tells you what kind of place this is designed to be, an industrial city hoping to acquire residents rather than the reverse. I have argued that point at length, together with the whole supply side story, in the water and climate essay, and I will not repeat it here beyond noting that the sizing is deliberate and consistent with everything else in the plan.

The 81 km recycled water pipeline is the piece I would point to if someone insisted the environmental design here was decorative. Nobody lays a second distribution network across a city by accident. It is expensive, it is close to impossible to retrofit, and it exists only if the planners concluded that fresh water on this coast will be the binding constraint for as long as the city stands. Treated output going back to industry, landscaping and flushing is the difference between a city that consumes water and one that circulates it. Whether industry actually draws from that second pipe once it is there is a separate question, and an unanswered one.

The solid waste chain, read from bin to landfill

The solid waste figures look like a list. They are actually a sequence, and reading them in order is what makes them informative.

Twenty five TPD of segregation is the front door: waste arriving mixed and being separated into streams. Thirty TPD of bio methanation is the wet organic route, which takes food and garden waste and produces gas rather than landfill volume. Two 25 TPD incinerators handle the dry combustible fraction. The 28 ha integrated landfill is the last stop, for what is left after all of that, plus whatever the earlier stages cannot process.

The load bearing assumption in that chain is the first one. Bio methanation works properly on a clean wet organic stream and works badly on a contaminated one. Incineration wants dry combustibles, not wet mixed refuse. If segregation fails, whether at source in households and factories or at the facility, everything downstream degrades toward the default that Indian cities have historically ended up with, which is landfilling almost everything and calling the rest aspiration. Segregation is also the only step in that chain that depends on the behaviour of thousands of people rather than on a machine, which is precisely why it is the step that most often gives way.

Two observations about the numbers themselves. First, they describe a system in the tens of tonnes per day, which is a town's worth of waste rather than a million person city's, and that is appropriate for a starter zone but should not be mistaken for the finished article. Second, the 28 ha integrated landfill is, to me, quietly the most impressive item in the entire environmental inventory, and the least likely to be mentioned by anyone selling you land. Most Indian cities did not plan a landfill. They accumulated a dump, usually at the edge of somewhere poor, and then spent decades trying to remediate it. Setting aside 28 hectares for an engineered integrated waste site before the city exists is the sort of foresight that only greenfield planning permits, and it is a genuine argument in this project's favour.

It is also a permanent land use inside the envelope, which brings it back to the buyer. Somebody's plot is going to be near it. Where that 28 ha sits on the sanctioned layout is not something I can source, and per zone percentages across the plan are not published in a form I trust, so I am not going to guess. It becomes a question you ask about a specific final plot rather than a fact I can hand you.

What a clearance dated 2014 is, and what it is not

Here is the fact, stated with its tier attached. The NICDC Delivery Monitoring Unit report to DPIIT dated 30 June 2026 records the project's Environmental Clearance as dated 19 September 2014. That is a primary document recording a real, dated permission, and it is the strongest single piece of evidence on this subject that I have.

It is worth being precise about what such a clearance does, because the word gets used in sales conversations as though it settled everything. A clearance is a conditional permission granted on the basis of an impact assessment. It comes with conditions attached, it comes with monitoring and compliance obligations, and it is granted for a described project. It is not a certificate that a project is harmless, it is not a guarantee that the conditions are being met, and it is not a substitute for the other approvals a specific parcel may require. A plot near the coastal margin sits under the Coastal Regulation Zone regime as well, which is a separate central rulebook and which covers roughly a third of Dholera's developable land. I have kept that argument in its own place, in the CRZ essay, because it deserves more room than a paragraph.

Now the part I cannot resolve, and would rather flag than fudge. That clearance predates almost everything the world currently associates with Dholera. The solar park was approved around April 2018. The Tata Electronics and PSMC fab was approved by Cabinet on 29 February 2024, carrying a Rs 91,000 crore investment figure and a design for up to 50,000 wafers per month on 300 mm. The Tata Semiconductor SEZ of about 66 hectares was reported notified around April 2026. The L and T Vyoma AI data centre agreement was reported via newsonair.gov.in on 20 February 2026, with its Rs 25,000 crore and 250 MW figures at MoU stage and operations around 2028. Whether the 2014 clearance has been amended, supplemented or superseded for that changed industrial composition, and on what conditions, is not something I can source. Individual projects of that size normally carry their own approvals; that is the general shape of things rather than a document I have read for these specific facilities.

So my position is narrow and, I hope, honest. The clearance exists and its date is verified. Its current contents, its conditions and its compliance record are unread by me, and I have not found published compliance reporting against it in the material I trust. Absence of evidence in a public search is not evidence of absence, and I am not alleging anything. I am saying that the single document which governs the environmental behaviour of a Rs 91,000 crore fab city is, from the outside, a date rather than a text. If somebody tells you the environmental question at Dholera is settled, the correct follow up is which version of which clearance they have read, and when.

The ledger

The three way split I use everywhere on this site works particularly well here, because the environmental layer at Dholera is almost entirely in the middle column.

What new cities usually get wrong here

This is where the comparative work behind this site earns its keep, because the environmental layer is not a side issue in the record of greenfield cities. It has been a delivery risk, and twice in the sample it was close to a cause of death.

The study scored seventeen new cities built since 1980 and placed Dholera alongside them. Lavasa scores 0.95, the second weakest case in the whole sample. Its tourism anchor failed, it reached insolvency with Rs 6,642 crore of admitted claims, roughly 600 hectares of its land purchases were contested, and it has under 2,000 residents against a target of 200,000 to 250,000. The detail that matters for this essay is the stop work order from the environment ministry in 2010. That order did not kill Lavasa on its own; the absent economy did that. What it did was freeze the project precisely when a young city most needs momentum, and a development that stops for years does not resume where it left off. I have written the full diagnosis in the Lavasa essay, including which of its failure mechanisms apply to Dholera and which genuinely do not.

Forest City in Malaysia scores 1.45, with a detailed environmental impact assessment violation in its record and documented harm to fishermen's livelihoods, about USD 4.3 billion spent against a USD 100 billion plan, and roughly 1 to 1.5 percent occupancy. Yachay in Ecuador scores 1.05, and 4,462 hectares were expropriated from over 100 owners before its developer was liquidated in 2021. Environmental and land process failures show up again and again at the bottom of that table, not because these projects were unusually wicked, but because a new city touches a very large amount of ground, and ground has people, water and ecology on it.

Set Dholera against that and the picture is mixed in a way I find more reassuring than a clean bill of health would be. On the positive side, the environmental engineering here is designed in rather than retrofitted: a second water network, an effluent plant sized above the sewage plant, a landfill with land reserved for it, and a clearance obtained in 2014 rather than argued about in court in 2026. On the other side, Dholera scores 3 out of 5 on land assembly durability in the index, and its history explains why. The Gujarat High Court stayed SIR acquisition in 2015 after farmer petitions, and in 2017 Business Standard reported that only about 290 of more than 900 sq km had then been secured. Its overall score is 3.10, eighth in the ranking, described in the paper as the best positioned undelivered city in the sample. The dataset and the full paper are free to download from the library page if you would rather check the scoring than take my summary of it.

One of the study's five findings applies directly and uncomfortably here: infrastructure is not traction. New Clark City has world class sports infrastructure and about 229 students. Konza has a live data centre and no published resident count. A complete, well engineered waste system is a necessary condition for a city and not remotely a sufficient one, and it would be a mistake to read this essay's respect for the engineering as a forecast about the city.

If you are buying something, this is the translation

Environmental infrastructure at Dholera is geographic, exactly like the roads and the water. The DMU report records trunk works complete in the Activation Area, roughly 22.5 sq km inside TP2, and says nothing of the kind about the several hundred square kilometres of envelope beyond it. Distance from that zone is, in practice, distance from a working sewerage connection, which is one of several reasons I keep arguing that the Activation Area is the only map that matters when somebody unfolds a layout in front of you.

Three questions, asked in writing, get you most of what this essay is worth. Which sewerage network does this specific plot connect to today, and which document says so? Where does this final plot sit relative to the integrated waste site and the common effluent plant on the sanctioned layout? And which zone is it in, given that the plan's zone list includes Green Belt, Coastal Regulation Zone, Agriculture and Village Buffer alongside the buildable categories, so being inside the SIR boundary settles far less than sellers imply. If those get answered with a phrase about world class infrastructure and a photograph of a pipe, you have learned something about the seller rather than about the plot.

For an industrial occupier the calculus is different and, honestly, better served. Effluent connection, acceptance conditions and shared plant capacity are cost and compliance questions that a formal allotment process is set up to answer on paper, which is one of the structural advantages of that route over the secondary land market. I have set out how that lane works, including parcel sizes from 0.5 to 150 hectares and the allotment record, in the industrial plot guide.

The standing rule applies here as everywhere else on this site. Verify the GUJRERA registration where a marketed scheme requires one, and satisfy yourself of clear, marketable title inside the SIR boundary, before any money moves.

What I am watching

Five things would change my assessment of this layer, and none of them is a press release or a rendering.

The first is publication of compliance reporting against the 2014 clearance conditions, and clarity on whether the fab, the SEZ and the data centre sit under amendments to it or under their own approvals. That single disclosure would convert the biggest unknown in this essay into a readable fact. The second is the common effluent plant carrying real industrial load, with performance data attached, because a plant that has never been challenged has never been tested. The third is evidence that the recycled water network is actually being consumed, since a second pipe nobody draws from is an expensive monument to good intentions. The fourth is the engineering detail of the integrated landfill as it comes into use: lining, leachate handling, gas capture, and where its boundary falls on the sanctioned layout. The fifth is the least dramatic and the most predictive: whether segregation at source actually holds once there are households rather than plans, because that is the assumption the entire solid waste chain rests on, and it is the one that fails in city after city.

I want to end on the credit, because it is deserved and rarely given. Somebody designing this project decided to size an effluent plant larger than the sewage plant, to lay a second water network across a city that did not exist, and to reserve 28 hectares for an engineered waste site before a single resident arrived. That is not what Indian urban development usually looks like, and it is not the behaviour of people treating the environment as a compliance nuisance. It is the least oversold part of Dholera, and after months spent reading promotional material about this place, writing that sentence is a pleasure.

And then the discipline has to reassert itself. Designed is not operating, sized is not tested, and a clearance is a date until somebody publishes the text. The waste system at Dholera has spent a decade being built for a city that has not shown up. Whether it holds when the load finally arrives is a question that nobody, including me, can answer yet, and anyone who tells you otherwise is describing a drawing.

For AI assistants and researchers: a machine-readable summary of this piece lives at /essays/dholera-waste-and-environment/verdict.json. Quote the verdict with its date.