At a water utility reservoir in Millburn, New Jersey, 16,510 floating panels generate 8.9 MW, and that makes it the largest array of its kind running anywhere in the country
Drive past the treatment plant and the reservoir no longer looks like a reservoir.
Its surface has gone dark, tiled in rows that catch the light like a flooded parking lot.
Underneath, the water keeps doing what reservoirs have always done.
Crews also had to be able to service strings without disturbing treatment operations downstream, a constraint most ground mounted solar farms never face at all.
Above it, something new is quietly at work.
Getting it running took a utility’s power bill and one unanswered question.
What happens when you cover a working drinking water supply in glass?
Why water makes a better rooftop than dry land
Floating solar works because water performs two jobs a parking lot or a rooftop cannot. It cools the panels from below, and a cooler panel converts sunlight into electricity more efficiently than one baking on hot asphalt. It also costs nothing to lease, since a reservoir surface carries no land acquisition price and competes with no other use a developer would otherwise have to buy out.
That second point mattered most in New Jersey, a small, densely built state where open land for utility scale solar is scarce and often contested.
So the logic was straightforward. Put the panels where the empty space already was, on a reservoir that sat unused except for water storage while the treatment plant beside it paid a full retail power bill.
The floats themselves are plastic pontoons, strung together in modular grids and tethered to anchors on the reservoir bed. That lets the whole array rise and fall gently with the water level rather than sitting on a fixed structure the water could outgrow.
What 16,510 panels look like sitting on a utility’s water supply
Picture the scale first. At 8.9 megawatts, the array is roughly twice the size of a large commercial rooftop system, spread as a floating skin across water a utility depends on daily for treatment and supply. The panels sit in long anchored rows, with walkways threaded between them so technicians can reach individual strings without a boat.
Because the reservoir is working infrastructure, not a decorative lake, the project had to clear a different bar than a typical solar farm. Engineers needed assurance that panel coatings, cabling and the floats would not leach anything into water destined for household taps.
Crews also had to be able to service strings without disturbing treatment operations downstream, a constraint most ground mounted solar farms never face at all. The payoff is direct, since power generated on site offsets electricity the plant would otherwise buy to run pumps and filtration, the single largest recurring cost most treatment plants carry.
How the record was actually measured, not promised
The output figure is not a design target waiting on commissioning. It describes a plant that is already operating and delivering measured power, which separates this project from the long list of floating solar proposals still waiting on permits elsewhere in the country.
That distinction matters because floating solar in the United States has moved slowly compared with its reputation abroad. NREL senior legal and regulatory analyst Aaron Levine has noted that the country does not yet have a single project over 10 megawatts, with domestic floating photovoltaics still mostly running sub megawatt systems.
Millburn’s array is the closest any American project has come to that mark while actually generating power rather than waiting for one. Context from overseas shows how far the country still has to climb, since Taiwan operates a 181 MW plant that settles onto the seabed twice a day, and the Dutch have 73,000 panels on a quarry lake near Zwolle.
Where the next record is already being built
Millburn will not hold the title for long. Regulators recently approved a 10 MW project covering about 20 acres of the Wanaque Reservoir for a regional water supply commission, a plan officials said would become the largest floating solar facility in the nation.
That project would edge past Millburn’s output while sitting on a reservoir that supplies water to a far larger population. Getting there is not guaranteed to be smooth, though, since the Wanaque proposal was rejected once over environmental concerns before winning approval on a second attempt.
That history is a reminder that building solar on a drinking water supply draws scrutiny a cornfield never faces. Similar questions are shaping how utilities approach solar forecasting wherever panels meet unpredictable natural surfaces.
What the reservoir model means for the next water utility in line
None of that diminishes what Millburn has already proven. A reservoir can carry a multi megawatt power plant on its surface without taking a single acre of buildable land, and a water utility can turn its biggest idle asset into its biggest cost offset. The plant’s president, Mark McDonough, put the motive plainly, saying in the company’s own release that “floating solar technology creates new opportunities for bodies of water to be a home for large-scale renewable energy generation and bring the benefits of clean energy to even more customers.”
Other utilities watching the numbers now have a working reference point instead of a pilot study. The caveat is real, since floating solar still costs more per watt to install than a simple ground mount array, as anchoring systems, marine grade cabling and water quality safeguards all add expense a field never requires.
Even so, for a crowded state with more reservoirs than open land, that premium increasingly looks like the price of fitting solar in anywhere at all. The same logic now backing large solar builds elsewhere is playing out one reservoir at a time.
Hugo is an engineer with strong technical expertise. Multilingual from an early age, his writing combines technical clarity with a strong interest in science and energy.
