Kali Gandaki A

Summary

Nepal's largest operating power plant of any kind (144 MW), the Kali Gandaki A run-of-river scheme at Mirmi has anchored the national grid since 2002. It is Nepali-owned (NEA), ADB/JBIC-financed, and a useful foil to the newer India-developed export projects: it shows what a multilaterally-funded, domestically-dispatched fleet backbone looks like — and also what two decades of wear do to one.

Specifications

Engineering

Three Toshiba Francis units in an above-ground powerhouse fed by a diversion dam with pondage. At 115 m gross head, Kali Gandaki A is a medium-head RoR design — less dramatic than Upper Tamakoshi at 822 m, but with pondage that smooths daily load in a way pure RoR plants cannot. The pondage is not storage; it does not shift energy between seasons. See Run-of-River Hydropower.

Output

The project was designed for ~860 GWh/yr but has averaged closer to 592 GWh — a ~30% shortfall that the ADB's own performance evaluation attributes to sedimentation, variable hydrology, and aging electromechanical equipment. This design-energy gap is a real-world demonstration of the principle in Installed MW, Energy, Seasonality, and Firm Power Are Routinely Conflated.

Financial

Built at USD 354.8 million with ADB and JBIC financing, Kali Gandaki A was for two decades Nepal's single largest plant — the reference case for multilaterally-financed domestic hydropower. Compare with Upper Tamakoshi (NPR 87 billion, domestic) and Chameliya Hydropower (USD 226 million for 30 MW, 11-year build) for the cost-efficiency spectrum.

Significance

Kali Gandaki A still matters out of proportion to its size. At 144 MW it is only about 4% of Nepal's 2025 installed fleet, but it was for two decades the single largest plant on the grid — and in most years it produced more firm, dispatchable energy than any other asset NEA operates. It remains the reference point when discussing what a Nepali-utility-run, multilaterally-financed project can look like on the buildability spectrum.

Compare Kali Gandaki A to two neighbours. Upper Tamakoshi (456 MW, 2021) is larger, newer and also Nepali-developed, but pure RoR without pondage. Arun 3 (900 MW, under construction) will be nameplate-bigger but peaking-design and India-dispatched. Kali Gandaki A sits between them in era and design philosophy — shared ADB lineage with Tanahu, NEA ownership like Upper Tamakoshi, and RoR-with-pondage shape that smooths evening load.

Limitations & Controversies

The design-energy gap. The project was designed for roughly 860 GWh/year but has averaged closer to 592 GWh — ~30% below design. The ADB's own performance evaluation flagged sedimentation, variable hydrology and aging electromechanical equipment as factors. That gap is a real-world reminder that Nepal's nameplate capacity is not the same as its delivered energy, a distinction central to Firm Power and Seasonal Mismatch analysis.

Rehabilitation overhang. In 2024 NEA floated a tender for design and supervision consultants to rehabilitate all three units. Any rehab of a fleet anchor plant introduces real derating risk during outage windows — exactly when the dry-season Storage Deficit bites hardest. Timing the rehab around Nepal's import / export contracts with India is a non-trivial coordination problem.

Pondage vs storage. Kali Gandaki A is not a reservoir project. The pondage it does have smooths daily load, not seasonal, so when monsoon flows fall off in January–May, Kali Gandaki A's contribution falls with them. Nepal's structural answer to that problem still sits in pages like Dudhkoshi Storage Hydroelectric Project, Tanahu Hydropower Project and Budhigandaki Hydroelectric Project — not here.

See also

• Gandaki Basin
• Upper Tamakoshi
• Run-of-River Hydropower
• Firm Power
• Nepal Electricity Authority (NEA)

Sources

• Kali Gandaki A — ADB Performance Evaluation & Wikipedia
• NEA Annual Report FY 2024/25
• NEA Transmission Annual Book 2077