Project Overview
Duration: 2025 - 2026
Contact: Dr Cheng Cheng, e: cheng.cheng1@anu.edu.au
Partners:
- The Ministry of Energy and Mineral Resources of the Republic of Indonesia
- Perusahaan Listrik Negara (PLN)
- Australian and in-country consultants
Funded by: Prospera: Australia Indonesia Partnership for Economic Development
Outputs:
- National pumped hydro site inventory (long list & short list)
- Five-stage shortlisting methodology and risk-ranking framework
- Pumped hydro shortlisting spreadsheet
- Pumped hydro background report
- Pumped hydro cost report
- Pumped hydro water requirement report
- Pumped hydro shortlisting report
- Ground-truth findings and reports
- Capacity building and stakeholder workshops
Introduction
Pumped hydro energy storage (PHES) has been identified by Indonesia as a strategic energy storage option. This project supported Australia’s Southeast Asian neighbour to scope options suitable for their unique energy transition requirements. Collaboration between Indonesia and Australia was facilitated through Prospera, a program built on more than two decades of Australian Government assistance that seeks to support Indonesian Government priorities.
In this project ANU researchers with expertise in engineering and social science worked together, drawing on the multi-award winning Pumped Hydro Energy Storage Atlases to map options nationwide. This body of work complements existing Indonesian pipeline planning serving to highlight opportunities beyond those already identified by Indonesian counterparts.
Working with in-country experts including partners from government, consultancies and the state owned electric power company PLN, ACES staff delivered a comprehensive and repeatable methodology for pumped hydro energy storage site identification and shortlisting. Indonesian agencies have been working with Prospera to explore the costs and policy needs in parallel projects.
Site identification
Selecting good PHES locations is largely about matching favourable topography with low-impact land use. In Indonesia, mountainous terrain and high rainfall make off-river options especially attractive because they can achieve large storage with relatively small inundation and avoid new barriers on major rivers. However, other configurations may be appropriate locally.
Most of Indonesia’s premium PHES sites lie in areas with sufficient rainfall for initial fill and are close to existing or planned transmission, which can improve feasibility and lower balance-of-system costs.
Methodology
The shortlisting methodology follows a five‑stage process designed to progressively screen a large initial inventory down to a manageable set of potential sites using technical and mappable considerations.
The staged approach has become feasible due to advances in data processing and mapping capabilities which enable systematic identification and comparison of very large numbers of potential reservoir configurations. As a result, modern PHES planning can begin with a much larger quantified inventory and then narrow down sites through progressively more detailed screening.
Stage 1 – Spatial analysis and an initial extensive list created through a detailed scanning of the terrain. Locations were identified where terrain allows reservoirs to be built in pairs. Potential sites were evaluated against technical factors and costed for ranking purposes (based on an ANU PHES cost model). This is the basis of the ‘Atlas’.
Stage 2 – Technical filters applied to create an extensive long list of PHES sites. Outcomes of Stage 1 include tens of thousands of potential reservoir pairs, with various configurations and in different cost classes. A set of technical filters (on head, reservoir separation, reservoir type, system size and cost class) ensure broad technical and economic suitability for Indonesia.
Stage 3 – Absolute ‘hard’ constraints registered. Beyond the built‑in Atlas layers, identified constraints in that landscape that must be avoided are added. Any overlap with potential sites leads to immediate exclusion from the shortlist. For example, overlap with areas with significant geotechnical risk.
Stage 4 – Further ‘soft’ constraints and risk ranking applied. Stage 3 sites are evaluated against a suite of social and environmental layers, identified as important. Each layer is treated as a constraint and a risk. An overall risk metric is calculated for each site overlapped by constraints. This stage narrows the shortlist.
Stage 5 – Non‑mappable factors are considered on the shortlist from stage 4. Issues that cannot be captured via reliable spatial, quantitative analysis require further investigation. Stage 5 includes feasibility studies and site checks on a narrowed small number of sites.
These stages gradually move from quantitative and mappable data toward locally informed expert assessment. The stages above allow for all possible sites to first be identified quantifiably (topologically) in a map and then for gradual elimination to narrow down to the most feasible sites.
A long list of over 185,000 potential reservoir sites was narrowed down to a shortlist of under 2000 sites using the 5-stage framework introduced above, which are further ranked based on a targeted risk-assessment framework.
Ground truthing
A team from ANU along with Indonesian counterparts undertook a visit to selected shortlisted sites in December 2025 to validate the appropriateness of the shortlisting methodology. This is known as ground truthing.
Following the ground-truthing trip, several updates were made. Key updates are listed below:
• That mapped risks often align with on-the-ground realities, but some discrepancies require qualitative assessment. Risks methods were refined to better reflect risks seen on the ground.
• The need for flexibility in reservoir pair design as the consideration of alternative reservoir pairings may be necessary, depending on the findings of the ground-truthing.
• That retrofitting existing reservoirs as PHES systems (i.e. ‘Bluefield’ PHES development) appears to be an attractive option due to the potential to have lower social and environmental impacts.
• That there is a need to plan for and integrate social, cultural, and environmental considerations from as early as possible in the shortlisting process. It is for this reason that the shortlisting process includes social, cultural and environmental risk factors and the methods include instructions for stages that come after shortlisting.
• The critical need to have a range of specialities involved in ground truthing. This is important because, for example, the process required consultations with people who understood the shortlisting process, people who understand dams and dam construction and with people who have the expertise to relate to the place and the people.
Knowledge sharing
Pumped hydro experts from the Australian National University delivered stakeholder and knowledge sharing workshops to serve as permanent learning resources for partner organisations. Written material accompanied the stakeholder workshops and technical resources were shared and adapted to ensure the site selection tool can be effectively used by Indonesian stakeholders at the completion of the project.
More about pumped hydro and the atlases
What is Pumped Hydro Energy Storage?
Pumped Hydro Energy Storage (PHES) is the dominant form of large‑scale electricity storage world‑wide, accounting for around 95% of global utility energy storage.
PHES uses two bodies of water at different elevations connected by pipes or tunnels. Water is pumped from the lower reservoir to the upper reservoir when electricity is plentiful and released downhill through turbines when power is needed. The round‑trip efficiency of PHES is roughly 80%.
PHES is a mature, long-duration storage technology that pairs well with variable renewable energy like wind and solar and provides grid services, black start (start-up that does not rely on an external grid), frequency/voltage support, and inertia.
From a system perspective, PHES and batteries increasingly compete with open-cycle gas turbines for flexible capacity. As solar and wind scale, storage helps address the temporal mismatch between supply and demand. However, even at lower renewable penetrations, PHES is a competitive provider of peaking and grid services.
Hybrid storage combinations (using PHES for cheap energy and batteries for cheap power) offer attractive flexibility. A low‑power, long‑duration PHES system (e.g. 150 GWh at 0.5 GW) coupled with fast‑response batteries can provide very low‑cost energy storage and ancillary services.
River based (open loop) PHES
Most of the world’s existing PHES fleet is river‑based and is integrated with conventional hydroelectric dams. The upper and lower reservoirs are part of a river valley and may be separated by a large distance. These open‑loop systems help smooth daily demand cycles, but new projects face limited suitable river sites and increasing social and environmental opposition.
Off-river (closed loop) PHES
Closed‑loop (off‑river) PHES exists where two small reservoirs are constructed away from major rivers and connected by a tunnel or penstock. Water is cycled repeatedly between the upper and lower reservoirs, allowing the same water to provide storage for decades with negligible losses.
Off‑river systems typically occupy only 1–50 km² of flooded area and can be sited almost anywhere in hilly terrain. They offer substantial advantages: better heads (vertical drops), reduced flood and sediment management costs, faster construction, and fewer environmental and social impacts compared with river‑based schemes.
Off-river pumped hydro has far lower inherent risk than river-based systems because there is no risk from rivers breaching dams during a 1000-year flood. Further, destruction of a dam wall due to water pressure is much less likely and impacts of destruction are much less likely to be severe since only relatively small volumes of water are stored in off-river PHES reservoirs.
The ANU Atlas identifies more than 600,000 off‑river reservoir pairs worldwide, demonstrating that PHES resource availability is not a constraint in most regions.
About the Pumped Hydro Atlases
The Pumped Hydro Energy Storage Atlases were created using Geographic Information System (GIS) techniques to identify potential pumped hydro energy storage sites worldwide.
The Atlases use the following categories to distinguish between various PHES sites
• A greenfield site refers to locations without existing water features.
• A bluefield site refers to locations with one or two existing reservoirs
• A brownfield site refers to locations with one or two existing mine sites
• And ocean refers to locations where the ocean acts as the lower reservoir.
• A seasonal site refers to locations where one reservoir is paired with a large river.
• A turkey’s-nest refers to locations where one of two reservoirs are entirely enclosed by dam walls, (e.g. a ring dam on flat land).
Ranking sites
Sites are ranked into AAA, AA, A, B, C, D, E classes based on factors including:
• Head - elevation difference
• Slope
• Water-to-rock ratio
• Energy storage capacity (GWh) and
• Duration of energy storage.
These features are mapped to an indicative cost using an internal cost model and classed from AAA (best) to E (Worst).
Class AAA and AA systems are characterised by large head (600-1600 m), large slope (>8%), large water-rock ratio (>8), large scale (150-1500 GWh) and long energy storage duration (50-200 hours).
Indicative cost rankings are designed for relative comparison between sites and are not intended as project-level cost estimates for individual sites.
Key siting considerations
At a high level, the factors that shape siting fall into five broad groups:
1. Physical & engineering geometry, such as head (vertical drop) and horizontal separation between reservoirs, basin shapes and slopes that influence tunnel length, dam wall volumes, inundation footprint, and cost/constructability. Larger heads and steeper conveyance generally mean more energy per hectare and smaller flooded areas.
2. Geotechnical and natural hazards, such as regional seismicity, proximity to active faults, volcanic hazard, and landslide susceptibility. These influence safety, operability, and the ability to insurance a site. Hazards may require buffers or route adjustments.
3. Environmental and ecological values, such as protected and conservation areas, as well as broader sensitivities such as key biodiversity areas, mangroves, non-protected forests, peatlands, riparian zones, and other habitats that warrant caution, avoidance, or mitigation.
4. Social and land use considerations, such as customary and community land, cultural heritage, human settlements, and productive agriculture (e.g., rice, mixed agriculture). These factors affect social licence, resettlement risk, consent processes, and the feasibility of offsets.
5. Infrastructure and access, such as proximity to strong transmission nodes, road access, and construction logistics. These shape deliverability and cost sharing with new renewables but are often optimised at design stage rather than used to reject otherwise strong geographies.