Australia’s 2GWh Kidston pumped hydro plant turns an old gold mine into a long-duration “water battery” and shows how storage, AI and site reuse can stabilise a renewable grid.
Australia has just switched on its first new pumped hydro plant in nearly 40 years – and it’s not just another power project. It’s a 2GWh “water battery” built inside an old gold mine, and it shows how green technology, smart markets, and AI-driven operations are starting to reshape the grid in very practical ways.
For anyone working in clean energy, infrastructure, or sustainability strategy, Kidston isn’t just an Australian story. It’s a clear example of how long-duration energy storage, data, and smart software can keep grids stable as coal exits and renewables scale.
This matters because the next decade in energy isn’t about adding more solar and wind – it’s about controlling them. Whoever gets storage and optimisation right wins on cost, resilience, and emissions.
What makes the Kidston pumped hydro project so important?
Kidston is Australia’s first new pumped hydro energy storage (PHES) facility to enter the National Electricity Market (NEM) in almost 40 years. It’s a 2GWh long-duration asset with 250MW of generation and 250MW of pumping capacity, built by repurposing two abandoned mining pits from the former Kidston Gold Mine in Queensland.
Here’s what stands out:
- 2GWh of storage: Enough to deliver 250MW for around 8 hours – ideal for evening peaks and multi-hour renewable gaps.
- Dual role in the market: It’s registered as both generator (KIDSPHG1, KIDSPHG2) and load (KIDSPHL1, KIDSPHL2) in AEMO’s Market Management System.
- 30-year commercial structure: EnergyAustralia has long-term dispatch rights through an Energy Storage Services Agreement, with an option to acquire the project.
- Mine-to-storage transition: Two disused mine pits now act as upper and lower reservoirs, turning a legacy fossil-era asset into green infrastructure.
From a grid perspective, Kidston is a flexibility anchor. It can:
- Soak up cheap or curtailed solar and wind when supply is high
- Release clean power when demand – and prices – spike
- Provide frequency regulation, voltage support, and other ancillary services that keep the NEM stable
The reality? This is exactly the sort of backbone infrastructure that high-renewables grids need, and it’s been missing from Australia’s toolkit for decades.
How pumped hydro fits into a green, AI-optimised grid
Pumped hydro is old tech, but it’s pulling new weight in green technology. In a modern grid, it works best when it doesn’t operate alone – it’s orchestrated by software, market signals, and increasingly, AI.
How the basic physics works is simple: pump water uphill when electricity is cheap, let it flow back down through turbines when electricity is expensive. But the value comes from when and how that happens.
With tools like advanced forecasting, machine learning, and optimisation platforms, operators can:
- Predict solar and wind output hours or days ahead
- Anticipate price spikes and market volatility
- Schedule pumping and generation to maximise both revenue and system support
Pumped hydro isn’t competing with batteries; it’s complementing them:
- Batteries excel at fast response (milliseconds), short durations (1–4 hours), and high cycling.
- Pumped hydro excels at long durations (6–24+ hours), big energy volumes, and providing inertia-like stability for the grid.
Most countries that are serious about deep decarbonisation will end up with a stack of storage technologies:
- Short-duration batteries close to load and renewables
- Medium-duration assets at substations and hubs
- Long-duration storage like PHES, compressed air, or flow batteries handling multi-hour and multi-day gaps
Kidston is a proof point that this stack isn’t theoretical – it’s being built, and the NEM is starting to operate like a properly optimised, software-driven clean energy system rather than a one-way fossil grid.
Queensland’s storage strategy: ambition, reality, and recalibration
Queensland has been both aggressive and pragmatic on pumped hydro, and Kidston sits right in the middle of that tension.
On one hand, the state has backed big projects like the 5.7GWh Borumba pumped hydro plant with AU$48 million in support. On the other hand, it has cancelled the proposed Pioneer-Burdekin mega-project, once touted as the world’s largest pumped hydro, over cost and environmental concerns.
This is the uncomfortable truth about large PHES:
- It’s capital-intensive
- It’s slow to build
- It can trigger real environmental and community pushback
Instead of betting everything on mega-schemes, Queensland has revised its energy roadmap toward a more balanced mix:
- Targeting 4.3GW of short-duration energy storage by 2030, mostly via batteries
- Progressing selected pumped hydro projects like Borumba and the 9.6GWh “Big G” project, which is still moving through regulatory approvals
I think this hybrid approach is the right call. Giant single projects look good in speeches, but a portfolio of flexible assets spreads risk, brings capacity online faster, and can be tuned to local conditions.
Kidston’s reuse of an old gold mine is a great example of that pragmatism:
- Reservoirs don’t need to be carved into untouched landscapes
- Existing topography and water assets keep environmental impacts lower
- Regulatory approvals are more achievable than with greenfield dams
For governments and developers outside Australia, the message is clear: long-duration energy storage is non‑negotiable, but the smartest path is modular, opportunistic, and grounded in real sites and real constraints.
Battery storage: the fast partner to slow, deep storage
While Kidston brings multi-hour pumped hydro back into the mix, battery storage is scaling across Australia at a rapid clip – and the two technologies are increasingly designed to work together.
A good example is Potentia Energy’s Quorn Park Solar Hybrid Project in New South Wales:
- 98MW of solar generation
- A 20MW/40MWh battery energy storage system (BESS)
- 16 battery units, each weighing ~27 tonnes
- Operations targeting 2026
The BESS there will:
- Smooth the solar output
- Support local grid stability
- Store excess solar during the day
- Release energy into the evening peaks when households and businesses actually need it
If you zoom out, you can see a pattern across Australia:
- Pumped hydro like Kidston handles deep, long gaps and system resilience
- Large grid-scale batteries support frequency response, firm renewables, and arbitrage
- Smaller behind‑the‑meter and community batteries take pressure off local networks
From a green technology standpoint, the interesting piece is that the intelligence layer is converging:
- The same optimisation engines can increasingly schedule both BESS and PHES
- AI tools can decide whether to charge the battery, pump water, or export power based on price forecasts and constraints
- Portfolio operators like EnergyAustralia can treat storage as a coordinated fleet, not a collection of one‑off assets
This is where AI stops being hype and starts acting like a real asset manager: it makes thousands of micro-decisions per day that humans simply can’t track in real time.
What Kidston means for businesses, investors, and policymakers
You don’t need to be building a pumped hydro plant to learn from Kidston. There are very practical takeaways for anyone involved in the green economy.
1. Long-duration storage is moving from concept to contract
For a long time, “long-duration energy storage” lived mostly in slide decks. Now we’re looking at:
- A 2GWh plant registered in the NEM
- A 30-year storage services agreement with a major retailer
- Integration into real‑world grid operations and market dispatch
If you’re a corporate offtaker or large energy user, this opens up new options:
- Structuring PPAs that are backed by long-duration storage
- Hedging price risk not just hourly, but across multi‑hour peaks
- Aligning ESG goals with genuinely firmed renewable supply
2. Repurposed infrastructure is a serious green technology strategy
Transforming an abandoned mine into a clean energy hub is more than a nice sustainability story – it’s a competitive advantage:
- Lower civil works cost than greenfield dams
- Faster permitting compared with untouched sites
- Strong narrative for investors focused on circular economy outcomes
Other sectors can learn from this mindset:
- Old industrial land becoming energy or data hubs
- Retired coal sites converted into battery or PHES projects
- Existing transmission corridors reused for new renewables
The companies that consistently spot these underused assets will out‑compete those that only look at pristine land.
3. Software and AI are the hidden multipliers
The raw physics of pumped hydro haven’t changed since the 20th century. What has changed is how assets are operated:
- Market optimisation platforms can increase storage revenues by double‑digit percentages by timing charge/discharge better.
- Grid-aware controls can provide multiple services simultaneously: energy arbitrage, frequency control, and voltage support from the same asset.
- Portfolio-level optimisation treats pumped hydro, batteries, and even flexible loads as one coordinated system.
If you’re planning any storage project in 2025 and beyond, and AI‑driven optimisation isn’t part of the business case, you’re probably leaving money on the table.
Where green technology is heading after Kidston
Kidston shows how green technology is maturing: less about shiny prototypes, more about integrated infrastructure that quietly does hard work in the background.
Australia’s storage build‑out – from 2GWh pumped hydro in a former gold mine to multi‑MWh batteries at solar farms and mines – is a preview of what many grids will need over the next decade:
- Flexible, long-duration energy storage as coal and gas retire
- AI‑orchestrated fleets of storage assets that stabilise renewables
- Smart reuse of industrial sites instead of expanding the footprint of heavy infrastructure
If your organisation is serious about net zero, resilience, or just avoiding volatile energy bills, the next logical step is clear: start treating storage, software, and site reuse as a single strategic package, not three separate topics.
Because the grids that thrive in a high-renewables future won’t be the ones that just built the most solar. They’ll be the ones that built the smartest combination of renewables, long-duration storage, and intelligent control – and Kidston is a strong sign that this future is now underway.