Inter-Basin Water Transfers and River Diversion Projects of Nepal

An inter-basin water transfer (IBT) moves water from a river basin with surplus flow (the donor basin) to one that runs short (the recipient basin), usually through a tunnel or canal cut across the watershed divide. Nepal's geography makes the idea attractive: its large perennial rivers — the Koshi, Gandaki and Karnali systems — are fed by Himalayan snow and keep flowing through the dry season, but they slice through the hills in deep gorges far from farmland. The Tarai plains, Nepal's grain basket, instead depend on smaller rain-fed rivers such as the Babai, Marin–Bagmati, Kamala, Tinau and Chisang, which shrink dramatically between November and May.

According to Water and Energy Commission Secretariat (WECS) estimates, Nepal's rivers carry roughly 225 billion cubic metres of water in an average year, yet most Tarai canal systems irrigate reliably only during the monsoon. Diversion projects take a modest slice of a big river's dry-season flow and feed it into an existing irrigation river, converting seasonal canals into year-round systems. Because the intake sits high and the outlet low, most schemes also pass the water through a powerhouse, adding hydropower as a by-product.

Two schemes — Bheri–Babai and Sunkoshi–Marin — are designated National Pride Projects (Rastriya Gaurav ka Aayojana) and are under construction, while Sunkoshi–Kamala, Tamor–Chisang and Kaligandaki–Tinau are at study stage. Nepal's first completed inter-basin transfer was built for drinking water: the Melamchi Water Supply Project, which brings the Melamchi River (Indrawati/Koshi basin) through a roughly 26 km tunnel into the Kathmandu Valley's Bagmati basin, delivering first water in March 2021 (Chaitra 2077 BS).

The Bheri Babai Diversion Multipurpose Project (BBDMP), headquartered in Birendranagar, Surkhet, diverts 40 cubic metres per second (cumecs) of the Bheri River — a major Karnali tributary — into the Babai River. Its headworks, a gated barrage at Chiple in Bheriganga Municipality of Surkhet district, feed a concrete-lined headrace tunnel about 12.2 km long (12,019 m finished length, 4.2 m internal diameter, per the project office). Exploiting a gross head of 151 m, the scheme will run three Francis turbines totalling 46.8 MW, with annual generation estimated at around 385 GWh, before releasing the water into the Babai.

The tunnel made engineering history: it was the first in Nepal excavated by a tunnel boring machine (TBM). A Robbins double-shield TBM operated by contractor China Overseas Engineering Group Co. (COVEC) broke through in April 2019 (Baisakh 2076 BS), roughly a year ahead of schedule — a milestone that changed how Nepali planners think about Himalayan tunnelling. The diverted water will provide year-round irrigation to 51,000 hectares in Banke and Bardiya districts through the Babai Irrigation Project's canal network.

Progress since the breakthrough has been slower. The headworks, powerhouse and hydromechanical packages lagged, and the government extended the completion deadline to fiscal year 2084/85 BS (2027/28 AD), about five years beyond the original target; cumulative physical progress stood at about 68.6 percent of the contractual target by the end of FY 2024/25. The project office puts the total cost at roughly NPR 36.8 billion.

The Sunkoshi Marin Diversion Multipurpose Project (SMDMP) transfers 67 cumecs of the Sunkoshi River — part of the Koshi basin — into the Marin Khola, a Bagmati tributary, through a 13.3 km tunnel. A 12 m high barrage on the Sunkoshi at Khurkot, on the Sindhuli–Ramechhap border, feeds the tunnel south to Kusumtar in Kamalamai Municipality-6, Sindhuli, where a 31.07 MW powerhouse (about 258.4 GWh a year, gross head 71 m, per NEA Engineering Company's detailed design) will discharge into the Marin. The augmented flow then enters the Bagmati Irrigation System, expanding year-round irrigation by about 122,000 hectares across Rautahat, Sarlahi, Mahottari, Dhanusha and Bara districts of Madhesh Province.

The project was declared a National Pride Project on 20 January 2020 (Magh 2076 BS). COVEC won the roughly NPR 10 billion tunnel contract in February 2021 and, using a 6.4 m diameter Robbins double-shield TBM — only the second TBM drive in Nepal — achieved breakthrough on 8 May 2024 (Baisakh 26, 2081 BS), about 11 months ahead of schedule despite fault zones and mixed geology of mudstone, sandstone, quartzite and granite.

The barrage and powerhouse have fared worse. The NPR 14.07 billion civil contract awarded in January 2023 to the Raman–Patel joint venture was terminated after physical progress stalled at about 11 percent with 60 percent of the contract period elapsed, and in May 2026 the Public Procurement Monitoring Office blacklisted the contractor for three years while the government moved to confiscate bank guarantees and re-tender the works. The project's July 2027 completion target is therefore under serious pressure even though the tunnel is essentially finished.

The Sunkoshi–Kamala Diversion is the largest Koshi-basin transfer on Nepal's drawing board and one of its oldest ideas: the 1985 Koshi River Basin Master Plan study supported by the Japan International Cooperation Agency (JICA) ranked a Sunkoshi-to-Kamala irrigation scheme among Nepal's top-priority projects, and diversion of Sunkoshi water to the Kamala is explicitly permitted under Nepal–India agreements on Koshi waters.

In the configuration described by the Saptakoshi High Dam Multipurpose Project and Sunkoshi Storage-cum-Diversion Scheme office — the joint Nepal–India investigation under which it is now studied — a diversion dam roughly 49 m high on the Sunkoshi would send about 72 cumecs through a tunnel of about 21.5 km (earlier studies proposed alignments as short as 16.6 km from a dam at Kurule) into the Kamala basin. A surface powerhouse of about 61 MW at Khadikhop would generate around 511 GWh a year, and a barrage at Uttarayani on the Kamala would command up to 175,000 hectares, initially in Siraha and Dhanusha and potentially extending to Saptari, Mahottari and neighbouring districts.

Madhesh Province governments have repeatedly demanded that the project be prioritised, since the Kamala basin suffers some of the Tarai's worst dry-season water stress. Because the scheme is bundled with the larger, diplomatically complex Saptakoshi High Dam study, it has no construction timeline yet, and Nepali water experts debate whether it should proceed as a standalone diversion the way Sunkoshi–Marin did.

The Kaligandaki–Tinau Diversion Multipurpose Project would divert the Kaligandaki River near Ramdighat in Syangja, with a diversion dam proposed at Pipaldanda in Rambha Rural Municipality of Palpa, sending flows through a tunnel of about 38 km into the Tinau River above Butwal. As envisioned when its project office opened in Butwal in June 2021 (Asar 2078 BS), it would irrigate about 107,000 hectares — roughly 54,000 ha in Kapilvastu and 53,000 ha in Rupandehi — and generate about 126 MW, at a cost initially estimated around NPR 138 billion (later estimates run higher). It remains at the study and detailed project report (DPR) stage.

It is also Nepal's most contested diversion proposal. Political parties, religious leaders and downstream communities in Gandaki Province have rallied against it, arguing that removing a large share of the Kaligandaki's winter flow would harm riverside settlements, downstream hydropower and the river's religious significance as the source of shaligram fossil-stones and the sacred confluences at Ridi and Devghat. The Gandaki provincial government has formally opposed the diversion, making inter-provincial water sharing a defining test case for Nepali federalism.

In the east, the proposed Tamor–Chisang Diversion (sometimes written Tamor–Chisapani) would tap the Tamor River, the Koshi's easternmost major tributary, and carry water through a tunnel of about 23 km to Letang in Morang district, feeding the Chisang Khola. Preliminary studies indicate it could irrigate on the order of 100,000 hectares in Morang and Jhapa and generate around 102 MW; it is promoted as the irrigation backbone for the eastern Tarai in the government's Agriculture Investment Decade (2024–2034) plans.

Nepal's diversion boom is inseparable from the arrival of tunnel boring machines. For decades, Himalayan tunnels were driven by drill-and-blast, with slow progress through the young, faulted rock of the Siwaliks and Mahabharat range — the Melamchi tunnel took the better part of two decades. Bheri–Babai's Robbins double-shield TBM changed the calculus in 2017–2019 by finishing a 12 km bore about a year early, and Sunkoshi–Marin's 6.4 m TBM repeated the feat in 2021–2024, breaking through 11 months ahead of schedule despite crossing two major fault zones. Both drives were executed by COVEC under Department of Water Resources and Irrigation (DoWRI) supervision.

The delivery lesson from both National Pride Projects, however, is that tunnels are no longer the bottleneck — surface works and contract management are. At Bheri–Babai the barrage and powerhouse have trailed the tunnel by more than half a decade; at Sunkoshi–Marin the dam-and-powerhouse contract collapsed and had to be re-tendered. TBMs de-risked the geology, but land acquisition, contractor capacity and payment cycles now determine when water actually reaches farmers' fields.

The case for inter-basin transfers is strongest in irrigation economics. Year-round water lets Tarai farmers move from one monsoon paddy crop to two or three crops a year across the districts that produce much of Nepal's food. The add-on powerhouses — 46.8 MW at Bheri–Babai and 31.07 MW at Sunkoshi–Marin, with larger plants proposed elsewhere — deliver relatively steady generation because design flows are guaranteed even in the dry season, complementing Nepal's run-of-river-heavy power system.

The risks sit mainly in the donor basins. Diverting dry-season flow reduces environmental flows downstream of the intake, affecting aquatic life, downstream users, existing and planned hydropower plants, and — acutely in the Kaligandaki's case — rivers of deep religious importance. Sediment-heavy Himalayan rivers also challenge intakes and turbine life, while Koshi-basin schemes such as Sunkoshi–Kamala intersect with Nepal–India treaty arrangements on downstream flows. Environmental impact assessments prescribe minimum releases, but independent monitoring of compliance remains a recurring civil-society demand.

On balance, Nepal's planners treat diversions as a middle path between building nothing and building giant storage dams: they use a small fraction of a large river's flow, avoid big reservoirs and displacement, and make existing canal systems perennial. How well Bheri–Babai and Sunkoshi–Marin close out their final contracts will largely decide how quickly the Sunkoshi–Kamala, Tamor–Chisang and Kaligandaki–Tinau proposals move from paper to construction.