Dholera Solar Park: The Real Numbers Behind the Megawatts

Every Dholera brochure I have read eventually plays the solar card. The wattage changes depending on who printed the thing: sometimes 1,000 megawatts, sometimes 5,000, sometimes a superlative I cannot trace to any document at all. The Dholera solar park is real, parts of it are genuinely impressive, and that is exactly why the inflation annoys me. When a true thing gets stretched, the reader who eventually checks the record ends up doubting the true part too, and this city cannot afford to lose the few claims it can actually prove.

So this essay does one job: it sorts the megawatts. Three different numbers circulate for the Dholera solar park, and they belong to three different categories of reality. Around 300 MW is commissioned and generating, built by Tata Power. A round 1,000 MW is the sanctioned Phase-I of the park. And 5,000 MW is a long-range ambition attached to a horizon around 2030. Sales copy collapses all three into one glowing figure. I am going to pull them apart, show where each one comes from, and then answer the question that actually matters: what does a solar park do for a city whose entire thesis is heavy industry?

My sources are the boring ones, which in this line of work is a compliment: the Central Electricity Authority's installed-capacity records, the NICDC Delivery Monitoring Unit report dated 30 June 2026, Tata Power's own disclosures, and the DSIRDA sanctioned development plan. Where a figure is a target rather than a fact, I will say so, because in Dholera the distance between those two words has historically been measured in years.

Three numbers that are not the same number

Here is the count, tiered the way I tier everything on this site: durable facts that survive checking against primary records, reported claims that trace to a source but not yet to hardware, and targets, which are intentions wearing a date.

The park was approved around April 2018, its Phase-I sanction sits with GPCL, Gujarat Power Corporation Limited, and the sanctioned development plan reserves a dedicated Solar Park zone inside the SIR, so the land-use foundation under the ambition is genuine planning rather than fantasy. But the distinction between the rows matters more than any single row. A sanctioned megawatt is a decision, while a commissioned megawatt is electricity. Only the second kind turns a meter, and a decade of Dholera marketing has been built on blurring precisely that line, across airports, trains, and megawatt counts alike.

What is actually generating: Tata Power's 300 MW

The commissioned piece is a Tata Power plant of roughly 300 MW, and its specifics are worth lingering on because they are the rare Dholera numbers that describe hardware you could stand in front of today: 873,012 modules across about 1,320 acres, mounted on single-axis trackers. Tata Power has described it as India's largest single-axis tracker installation, which I treat as a company claim rather than a verified superlative, but the capacity itself is durable. It shows up in the Central Electricity Authority's records as of September 2025 and again in the NICDC Delivery Monitoring Unit report of 30 June 2026, which is the closest thing this city has to an audited progress ledger.

Single-axis tracking, for readers who have not stood in a solar field, means the long rows of modules pivot slowly from east to west through the day, following the sun roughly the way a sunflower does. The gain over fixed frames is real energy, not cosmetics, and on the Bhal it is the obvious engineering choice: this is flat, low-lying, sun-hammered land with barely a tree to cast a shadow. The same flatness that gives Dholera's planners a documented drainage problem is close to ideal for laying out module rows by the hundred thousand.

I want to be precise about what this plant proves and what it does not. It proves the solar park is not vapour: a large, grid-connected, generating power asset has been operating at Dholera through years in which much of the rest of the city existed mainly as a schedule. It does not prove the park is finished, and it does not mean the number on the brochure is installed. Think of it as the down payment, not the house.

The missing 700 megawatts

Subtract the built from the sanctioned and you get the interesting number: about 700 MW of Phase-I remains under development. The stated target for completing it is March 2027. The documented reason it is late is a set of tariff disputes that went before Gujarat's electricity regulator, GERC.

A tariff dispute sounds like paperwork and is actually existential. A solar plant is a decades-long bet whose economics hang almost entirely on the price per unit someone has signed to pay for its output. When the generator and the buyer stop agreeing on that price, financing hesitates and construction slows, because a plant built on a contested tariff is an expensive way to own an argument. I do not have the case files to tell you which side is right, and I will not pretend otherwise. What the record supports is narrower and more useful: the balance of Phase-I is real enough to be fought over, and the fight has cost it time.

How should you read March 2027, then? The same way I read every date this city publishes: as a claim with a track record attached. I wrote a full essay on reading the airport's dates, and the method transfers without modification. The park was approved around April 2018; eight years on, Phase-I stands at less than a third of its sanctioned capacity. That history does not make March 2027 false. It makes it a target, and targets earn belief from commissioning records rather than press coverage. When the CEA tables show the number climbing, I will upgrade my language. Until then, the 700 MW sits in the pending column, where honest accounting keeps it.

The 5,000 MW figure, filed where it belongs

Now the big one. You will meet 5,000 MW in nearly every sales deck that mentions Dholera and electricity, usually phrased as though the capacity hums away today. It does not exist. The 5,000 MW figure is a long-range ambition with a horizon around 2030. It is not sanctioned as a programme, and it is certainly not installed. What exists at that scale is intent: the master plan reserves the zone, and the state repeats the number with conviction.

I am not mocking ambition; cities are built out of it, and a 5,000 MW park on this much flat, empty land is not physically absurd. But claims need a filing system, and the drawer this one belongs in is labelled target. My working rule, argued at length in the essay on completion dates, is that an ambition becomes a programme when sanctions and money attach to it, and becomes a fact when the CEA logs it. By that rule the park today reads: 300 MW fact, 1,000 MW programme, 5,000 MW aspiration. Anyone who quotes the third number without the first two is selling, not describing.

Why an industrial city needs a firm power story

Here is where I part company with the boosters and the cynics at the same time, because the interesting question is not how many megawatts the park has. It is what kind of electricity a fab city actually needs, and solar alone is not the answer to that question anywhere on earth.

The anchor of Dholera's entire thesis is the Tata Electronics fab: Rs 91,000 crore of committed investment, cabinet-approved on 29 February 2024, targeting first silicon around December 2026, with commercial production reported for mid-2028 and a design capacity of up to 50,000 wafer starts a month when fully ramped. Semiconductor fabs are among the most power-sensitive factories humans build. They run around the clock, every day of the year, and even a momentary voltage sag can scrap hours of work in progress. A fab does not want cheap power at noon and a prayer at midnight. It wants firm, clean, uninterrupted supply, permanently.

Solar, on its own, is the opposite temperament: generous at noon, absent at night, moody through the monsoon. So let me say the unfashionable thing plainly: the Dholera solar park does not power the fab in any meaningful engineering sense, and nobody serious ever claimed it would. Firmness comes from the grid, and the grid inside the SIR is its own quietly serious story. The Activation Area's trunk build includes three 66 kV substations and about 115 km of underground power duct, according to the planning and delivery records, and the DMU report of 30 June 2026 records the trunk works as complete. I walked through that inventory in the Activation Area essay, and to a factory owner it matters more than any tracker field.

What solar gives an industrial city is different and still valuable. It cheapens the average electron. It hands export-facing manufacturers a renewable share to report, which has quietly moved from marketing garnish to procurement checkbox in electronics, because chipmakers' customers carry climate commitments that flow down the supply chain. And it signals to exactly the tenants Dholera wants that the intent here is energy surplus, not energy scramble.

The demand side is assembling as well. The L&T Vyoma AI data centre agreement was signed on 20 February 2026; the agreement itself is well documented, while the reported Rs 25,000 crore and 250 MW figures remain MoU-stage numbers in my ledger, with operations discussed for around 2028. Above that sits the state's Viksit Gujarat Data Centre Policy 2026-29, which names Dholera the primary cluster in a statewide ambition of 7.5 GW of data centre capacity. That is a state target, not capital committed to Dholera, and I would distrust any pitch that quotes it as committed money. But targets shape where wires get built, and wires shape where factories land. Fabs and data centres share one appetite, which is enormous, unbroken power, and a city planning to host both has to tell a credible generation story. A working 300 MW plant plus a sanctioned gigawatt is the credible core of one.

How solar fits the fab-city thesis

Regular readers know my frame: the fab is the ballgame, and everything else in Dholera either supports a working industrial city around that anchor or it is decoration. Run the solar park through that filter and it comes out better than I expected when I started checking, for three reasons.

First, chronology. The plant's roughly 300 MW was in the CEA's records by September 2025, before the expressway opened on 31 March 2026, before the semi-high-speed rail was approved in May 2026, and while the airport was still hunting for its opening date. For years, the strongest honest rebuttal to "nothing is happening in Dholera" has had this plant on the list. The full rebuttal as of August 2026 reads: fab civil work reported past the halfway mark by mid-2026, 300 MW of solar generating, the expressway open, trunk infrastructure recorded complete. The solar park is the senior member of that list, the one that was producing something measurable while everything else was still a date.

Second, coherence. A semiconductor cluster sitting beside a large renewable park is a story that holds together for the people who will one day audit it. I will not pretend a chip buyer chooses a foundry because of a tracker field on the Bhal. But renewable content in the supply chain has become a question customers actually ask, and Dholera's version of the answer is at least made of steel and silicon rather than adjectives.

Third, proportion. Solar is the one Dholera asset class where the true number is already large enough that nobody needs to inflate it. It has not stopped anyone, of course, which brings me to the practical part.

A watchlist, and a word on plot pitches

If you want to track this park without trusting me or anyone else, watch five things. First, the Central Electricity Authority's installed-capacity tables: Dholera's commissioned figure either climbs above roughly 300 MW or it does not, and that ledger is indifferent to press releases. Second, March 2027: whether the remaining 700 MW arrives on target, arrives in tranches, or slips again, and by how much. Third, movement at GERC on the tariff disputes, because the concrete follows the ruling. Fourth, any formal sanction beyond the 1,000 MW Phase-I, which would be the day 5,000 MW begins its journey from ambition toward programme. Fifth, the next delivery-monitoring reports in the DMU's line, which is where a claimed megawatt goes to become a recorded one. I keep the running ledger across all of Dholera's systems in the current status audit, updated as the documents move.

And because everything in Dholera eventually becomes a land pitch, one caution. You will hear the solar park quoted as a reason plot prices must rise. Hold that argument to the same standard as the megawatts. A power plant nearby is a genuine piece of a city thesis and a useless substitute for diligence on the specific parcel in front of you. If money is about to move, verify the scheme's GUJRERA registration and insist on clear title inside the SIR boundary before anything else, because megawatts do not launder a bad title.

The honest version of the Dholera solar park story is still a good story, which in this market makes it nearly exotic. Around 300 MW is generating, and that is a fact I can stand on. Another 700 MW has a target date and a documented fight to get through first. And 5,000 MW is a hope with a zone reserved for it. Quote all three tiers in that order and you will be more accurate than most of the industry that sells this city for a living.

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