Two hundred and fifty megawatts, up to eight hours’ worth of it, sitting in a hole in the ground that used to produce gold. That’s the pitch for Kidston Pumped Storage Hydro, and after years of slipping dates it is finally in the phase where the numbers start to matter more than the promises: commissioning. Water is moving between the two old mine pits, the pump-turbines are being run up and tested against the grid, and Genex Power’s engineers are doing the unglamorous work of proving a decades-old open-cut gold mine can behave like a giant chemistry-free battery.
I drove out that way a few years back, past Mount Surprise, on a story that had nothing to do with energy at the time — the kind of long, flat, empty run where the horizon does all the talking and you start doing sums in your head just to stay entertained. Kidston itself sits near Georgetown in the Etheridge Shire, a good few hundred kilometres inland from Cairns, which tells you something about why this project has always been as much about transmission as it is about water.
What’s actually in the ground at Kidston #
The site is the old Kidston gold mine, closed since the late 1990s, and its two open-cut pits — Eldridge and Wises — have been repurposed as the upper and lower reservoirs. Water is pushed uphill when the grid has cheap surplus power (typically the rooftop-solar-soaked middle of the day) and released back downhill through the same tunnels to spin turbines when the grid needs it, usually the evening ramp. It’s the same physics as Tumut 3 or Wivenhoe, just built inside somebody else’s old hole rather than a purpose-dug valley.
The rated output is 250MW, and the storage is commonly quoted at around 2,000MWh. Those two numbers get conflated constantly, so let’s be precise about what they mean, because this is exactly the kind of distinction that matters when you’re comparing pumped hydro to a battery.
- 250MW is what the plant can deliver at any single instant — the power rating.
- 2,000MWh is the total energy in the tank — how long it can sustain that instant before running dry.
- 2,000MWh divided by 250MW gives roughly eight hours of full-rate discharge, which is the duration figure that actually separates Kidston from most grid-scale batteries currently operating in the NEM.
Compare that with a large lithium-ion battery like the Victorian Big Battery, rated around 300MW but only about 1.5 hours of duration at full output. Different tool, different job. I’ve made this point before and I’ll keep making it: the industry conversation still slides too easily between MW and MWh as if they’re interchangeable, and they’re not — a plant’s worth to the grid depends on both. Our own piece on pumped hydro versus big batteries for firming goes through this trade-off in more detail, and it’s worth reading alongside the discussion of whether big batteries are being built for the wrong job in the first place.
The bill, and who’s paying it #
Kidston’s costed history is not a clean line. Early estimates for the pumped hydro stage sat well under the number now being talked about; construction realities — tunnelling through variable rock, wet-season disruption, contractor scheduling — pushed the total cost up over time, and Genex has flagged cost growth in its own ASX filings rather than pretending otherwise. Let’s be careful with that number rather than repeat an old headline figure as if it’s still current: the position has moved since the project’s original costings, and readers chasing a precise final dollar figure should go to Genex’s own market disclosures rather than take an older press estimate as gospel.
What’s clearer is the funding stack. The Northern Australia Infrastructure Facility provided a concessional debt facility specifically for the pumped hydro build, ARENA and the CEFC backed earlier feasibility and the co-located solar stage, and Japan’s J-Power became a major shareholder in Genex, later increasing its position — a sign that at least one large offshore utility was prepared to underwrite construction risk on an Australian pumped hydro project when local capital markets were still cautious about the asset class. Queensland’s state government has also had skin in the game through its broader pumped hydro ambitions, laid out in the Queensland energy plan’s pumped hydro and public ownership push — though it’s worth separating that state-owned pipeline, built around Borumba and Pioneer-Burdekin, from Kidston, which remains a privately developed asset with public financing support rather than public ownership.
Why the timeline kept slipping #
Commissioning was originally pencilled in for around 2024. It didn’t happen then, and it’s worth being honest about why rather than waving it away as generic “construction delays”. Underground works of this scale — tunnels linking two mine pits with a significant head difference, an underground powerhouse, penstocks — carry geological risk that doesn’t show up until you’re actually boring through it. North Queensland’s wet season adds a genuine seasonal constraint on civil works that a project in, say, the Snowy region doesn’t face in the same way. And a project built by repurposing existing infrastructure sounds cheaper on paper than it often turns out to be in practice, because you inherit somebody else’s geology, somebody else’s pit walls, somebody else’s assumptions.
None of that is unique to Genex. Snowy Hydro’s own pumped hydro build has become the reference case for how badly a tunnel-heavy project can blow out on time and cost, and Kidston’s slippage has been modest by comparison. But it’s a reminder that pumped hydro, for all its appeal as long-duration storage, is closer in construction risk profile to a tunnelling project than to a battery installation you can bolt together in a paddock in eighteen months.
What AEMO actually has it down as #
AEMO’s Integrated System Plan tracks committed and anticipated generation and storage projects for the National Electricity Market, and Kidston has moved through those categories as construction has progressed — from anticipated to committed as financing and construction milestones were locked in. That distinction matters more than people give it credit for: AEMO’s planning documents don’t just list projects that developers hope to build, they weight capacity by how confident AEMO is the thing will actually turn up when needed, which feeds directly into how the market operator plans for reliability in north Queensland. Readers who want the mechanics of how that dispatch and planning process actually works day to day should have a look at our explainer on how the NEM dispatches power every five minutes.
North Queensland specifically has been short of dispatchable, flexible capacity for a while, leaning on an ageing fleet and long transmission runs back to the southern load centres. A 250MW/2,000MWh asset that can absorb the region’s considerable rooftop and utility-scale solar in the middle of the day and give it back in the evening is exactly the shape of asset the region’s stability case has been missing.
The renewable hub it’s supposed to anchor #
Kidston was never meant to be a standalone plant. Genex has built it alongside an existing solar stage on site and a wind development, with the pumped hydro intended as the firming backbone for the whole hub — soak up the solar and wind output when it’s abundant, release it when it isn’t. That’s a coherent design on paper. Whether it plays out that cleanly in the market depends on wholesale price patterns that have themselves been shifting, including the now-familiar pattern of negative prices at midday squeezing the economics of exactly the charging window pumped hydro is meant to exploit.
My read on it #
Here’s my mildly unfashionable view: I think the industry commentary around Kidston has spent too long focused on the delays and not enough on the fact that eight-hour duration storage is genuinely rare in this market, and rarer still at this scale outside the existing Snowy and Tasmanian hydro fleets. Batteries are winning the short-duration arbitrage game comfortably and will keep doing so. But north Queensland’s evening peak doesn’t care how fashionable lithium-ion is; it cares whether something can deliver 250MW for six or seven hours running, and right now Kidston is one of a very small number of assets anywhere in the NEM that can plausibly do that once it’s through commissioning cleanly.
The next milestone that actually matters isn’t a press release, it’s AEMO’s own registration data showing the plant contributing to the market at scale across a full evening peak, more than once, in more than one season. Until that shows up in the public dispatch record, I’d treat “commissioning” as a status, not a finish line.
— Anjali Rao, Grid & Storage Correspondent
Image: Collywolly (CC BY-SA 4.0) via Wikimedia Commons