Can River Interlinking Cut Monsoon Rain and Starve the Sundarbans
A river can be moved on a map long before its consequences appear in the sky.
That is the central warning emerging from two peer-reviewed findings on India’s river interlinking ambitions. One looks upward, at the monsoon. The other looks downstream, at the delta. Together, they challenge the comforting idea that water can be shifted from “surplus” basins to “deficit” basins without changing the climate and ecology that made those basins in the first place.
A study published in Nature Communications by scientists at IIT Bombay, IIT Madras and the Indian Institute of Tropical Meteorology modelled what could happen when large volumes of surface water are carried into previously unirrigated regions. Their finding is stark: more irrigation means more evaporation, and that can alter land-atmosphere feedbacks enough to potentially cut September monsoon rainfall by up to 12 per cent in India’s driest basins.
These are the very regions river interlinking is meant to protect from drought.
A separate study by researchers at the University of Colorado found another risk. Interlinking could reduce silt reaching the Ganga-Brahmaputra delta by around 30 per cent, with discharge falling in 24 of the 29 rivers assessed. That silt is not waste. It is the raw material of the Sundarbans, the world’s largest mangrove forest, a UNESCO World Heritage Site shared by India and Bangladesh.
The two findings point to one lesson: rivers do not end at diversion points, reservoirs or state borders. They continue into clouds, estuaries, mangroves, farms, fisheries and foreign policy.
The promise of river interlinking rests on a dangerous simplification
India’s river interlinking project has long been sold through a simple narrative. Some basins flood, others dry out. Move water from one to the other, and the country gains both flood control and drought relief.
The attraction is clear. India faces severe water stress, uneven rainfall, rising irrigation demand and recurrent floods. A national-scale engineering plan offers a sense of command over a problem that often feels chaotic. It promises storage, transfer, predictability and political certainty.
Yet the hydrology behind that promise is not simple.
The idea of “surplus” and “deficit” rivers assumes that water sitting in a channel can be counted separately from the wider system that produces it. It treats a river mainly as a delivery mechanism. That view may work for a canal command area or a reservoir balance sheet. It becomes fragile when applied to the monsoon.
The monsoon is not just rain falling into rivers. It is a seasonal circulation shaped by temperature gradients, soil moisture, vegetation, evaporation, aerosols, topography and ocean conditions. Land does not passively receive rainfall. It helps condition the atmosphere above it.
When river interlinking brings water into previously unirrigated land, it changes what that land sends back into the air. Wet soils and irrigated crops evaporate more water. That moisture can cool the land surface. Cooling can then weaken the thermal contrast that drives atmospheric circulation. Under some conditions, this can reduce the very rainfall the project is supposed to manage.
This is where the Nature Communications finding matters. The researchers did not merely ask whether canals can move water. They asked what happens after that water changes the land surface. Their modelling suggests that large-scale irrigation associated with interlinking could reduce September rainfall by as much as 12 per cent in the driest basins.
September matters because it sits near the retreating phase of the southwest monsoon. Rainfall during this period can decide soil moisture, reservoir levels and crop outcomes after the main monsoon pulse. A reduction in already dry basins is not a marginal side effect. It could undermine drought resilience.
This is the first part of the uncomfortable arithmetic. A plan designed to reduce drought may change land-atmosphere feedbacks in ways that deepen rainfall stress in drought-prone regions.
That does not mean every canal causes a failed monsoon. It means the scale of intervention matters. Local irrigation and national-scale river transfer are not the same class of action. When a project seeks to rearrange water across basins, the atmosphere becomes part of the project area.
Policy has not caught up with that fact.
The atmosphere is part of the river basin
The older language of river planning often imagines basins as bounded units. Water falls, flows, is stored, diverted and consumed. The accounting follows the channel.
Climate science widens the boundary.
A basin is also connected to the air above it. Evapotranspiration from fields, forests and wetlands can influence humidity, temperature and circulation. Soil moisture affects how much incoming energy becomes heat and how much becomes water vapour. Crops with different rooting depths, irrigation schedules and surface properties change the exchange between land and atmosphere.
Large-scale river interlinking would not simply move water from Point A to Point B. It would create new wet surfaces in dry areas, alter crop choices, change irrigation intensity and shift seasonal evaporation patterns. Those changes can feed back into rainfall.
This is especially sensitive in a monsoon system. The Indian summer monsoon depends on pressure gradients and heat contrasts over land and sea. If widespread irrigation cools land surfaces, it can weaken upward motion and moisture convergence in certain regions. The details are complex, but the core insight is clear: water transfers can change rainfall, not just respond to it.
That insight should reshape how India evaluates river interlinking monsoon impact, Sundarbans sediment loss, India Bangladesh water dispute, Ganga Brahmaputra delta, climate science river linking as one connected policy problem rather than separate files in separate ministries.
The atmospheric risk also challenges the moral framing of the project. River interlinking is often presented as a national duty to bring water to drylands. That argument carries weight in a country where drought damages livelihoods and deepens rural distress. But a drought-relief project must be judged by whether it actually reduces drought risk under real climate conditions.
If the modelling shows possible rainfall suppression in dry basins, the project cannot be defended only through canal capacity or projected command area. It has to answer harder questions.
What happens to local rainfall after new irrigation changes evaporation?
Which regions gain water on paper but lose rainfall in practice?
How do these changes interact with warming, heatwaves and groundwater depletion?
Who bears the risk if the monsoon response differs from engineering expectations?
These are not anti-development questions. They are development questions. A canal that ignores the atmosphere is not ambitious. It is incomplete.
The delta also pays for upstream decisions
The second warning comes from the other end of the system.
The University of Colorado study found that river interlinking could reduce silt reaching the Ganga-Brahmaputra delta by around 30 per cent. It also found discharge dropping in 24 of the 29 rivers assessed. In a delta, those numbers are not abstract.
Deltas survive by receiving sediment. Rivers carry sand, silt and clay from upstream catchments. As flow slows near the coast, sediment settles, builds land, nourishes wetlands and helps offset subsidence and erosion. In a tide-dominated, cyclone-prone delta such as the Ganga-Brahmaputra, this sediment supply is a form of protection.
The Sundarbans depend on it.
The mangrove forest sits where fresh water, sediment and tides meet. Its islands and mudflats are not fixed platforms. They are living geomorphic features, built and rebuilt by water and silt. Mangrove roots trap sediment. Sediment raises land. Raised land helps mangroves survive tidal flooding. Mangroves then reduce wave energy and provide shelter for people and wildlife.
Cut sediment, and the system weakens.
A 30 per cent reduction in silt reaching the delta could affect channel stability, island formation, erosion patterns and the ability of mangrove areas to keep pace with sea-level rise. It could also change turbidity, nutrient flows and habitat conditions for aquatic species. The exact response would vary across the delta, but the direction of concern is obvious.
The Sundarbans are already under pressure from sea-level rise, cyclones, embankment failures, aquaculture expansion, salinity and human settlement stress. Less sediment would remove one of the few natural mechanisms that helps the delta defend itself.
This is why the phrase “starve the Sundarbans” is not rhetorical excess. Sediment is food for a delta. Without it, wetlands sink, shorelines retreat and protective mangrove belts become harder to sustain.
The problem is not only ecological. It is human.
Millions of people in the wider delta depend on fisheries, agriculture, forest resources and storm protection. When freshwater flow declines and salinity rises, drinking water sources suffer. Crops become harder to grow. Fish nurseries shift. Embankments face more stress. Cyclones arriving from the Bay of Bengal meet a landscape less able to absorb their force.
Bangladeshi scientists have raised this alarm directly. Reduced upstream flow could lower discharge at the Farakka Barrage, accelerate salinity intrusion into coastal rivers and destabilise a delta that already absorbs repeated cyclone impacts. Their concern is not a distant environmental objection. It is grounded in the lived geography of a downstream country.
The delta receives the consequences of upstream certainty.
The India-Bangladesh dimension cannot be treated as an afterthought
The Ganga and Brahmaputra do not end at India’s borders. They cross into Bangladesh before reaching the Bay of Bengal. That fact should be central to any basin-scale planning.
Yet India has never formally consulted Bangladesh on the river-linking project’s transboundary basins. This is a serious gap. A national project that changes flows in shared rivers is not only domestic infrastructure. It is a regional water decision.
The India-Bangladesh water relationship already carries historical weight, especially around the Ganga and the Farakka Barrage. Downstream concerns over dry-season flow, salinity and navigation have shaped diplomacy for decades. Adding large-scale inter-basin transfers without formal consultation risks turning a technical project into a trust deficit.
International river governance rests on a basic principle: upstream states should not cause significant harm to downstream states, and shared rivers require prior information and consultation. The practical form of that principle varies across treaties and institutions, but the norm is widely recognised.
For Bangladesh, the stakes are high because the delta is the country’s physical foundation. Changes in upstream discharge do not remain in river gauges. They appear in salinity lines, crop choices, drinking water stress, fisheries, embankments and cyclone vulnerability.
For India, the stakes are also high. Ignoring downstream risk can create diplomatic strain, legal criticism and reputational damage. It can also lead to technically weaker planning, because downstream data and delta science are essential to understanding the full system.
A project of this scale needs transboundary modelling, not just transboundary reassurance.
That modelling should include:
Seasonal flow changes in the Ganga, Brahmaputra and their distributaries
Sediment transport across dry and wet seasons
Salinity intrusion under sea-level rise and cyclone storm surge
Delta subsidence and erosion
Mangrove health, fisheries and livelihoods
Atmospheric feedback from expanded irrigation
Climate change scenarios, including extreme rainfall and drought years
The central question is not whether India has the sovereign right to build infrastructure. The question is whether infrastructure on shared rivers can be legitimate if it excludes the downstream country from the science that justifies it.
A project sold as flood and drought control could export both problems abroad while creating new ones at home. That is the worst kind of regional planning: expensive, confident and incomplete.
The science is catching up with warnings that were once dismissed
Critics of river interlinking have warned for decades that the project treats rivers as plumbing. They argued that basin transfers could disrupt ecology, displace communities, weaken downstream flows and fuel interstate or international conflict.
For a long time, these concerns were easier to dismiss than engineering diagrams. A canal looks concrete. A feedback loop looks uncertain. A reservoir has a capacity figure. A mangrove’s geomorphic dependence on sediment is harder to fit into a ribbon-cutting speech.
That imbalance is changing.
The newer studies do not merely repeat old fears. They sharpen them. They put numbers on pathways that critics had described in broader terms. One study estimates a possible rainfall reduction in dry basins through land-atmosphere feedback. Another estimates sediment loss to the delta and discharge reduction across many rivers.
This does not mean the models are final truth. All models carry assumptions. They simplify reality. They depend on input data, scale and scenario design. Climate and sediment systems are difficult to simulate, especially under future warming.
But uncertainty cuts both ways. It does not justify pressing ahead with irreversible infrastructure. If anything, uncertainty raises the burden of proof.
A responsible project would treat these findings as a reason to pause, revise and subject the plan to independent review. It would ask whether the intended benefits survive when atmospheric feedback, sediment budgets, downstream rights and climate change are included. It would make data public. It would invite scrutiny from Indian and Bangladeshi scientists. It would compare river interlinking with less risky alternatives.
Those alternatives exist. They may be less dramatic than national water grids, but they are often closer to the sources of water stress.
Restoring tanks, ponds, wetlands and floodplains
Improving irrigation efficiency without expanding water-intensive cropping in dry regions
Protecting groundwater recharge zones
Changing procurement and crop incentives that drive thirsty crops into unsuitable areas
Reducing leakage and inequity in existing canal systems
Using basin-level floodplain zoning instead of only embankments and diversions
Rebuilding local water institutions that manage scarcity and seasonal variability
None of these options is easy. Some require politically difficult reform. But they do not assume that the monsoon, the delta and a neighbouring country will quietly absorb the side effects of a grand transfer scheme.
This is where inspiration matters. South Asia does not lack engineering ambition. It needs ecological intelligence equal to that ambition.
The region has world-class climate scientists, hydrologists, sedimentologists, delta researchers, disaster experts and social scientists. It has communities that have managed floods, tanks, wetlands and tidal rivers for generations. It has hard-earned experience with cyclones, drought and river erosion. A better water future can draw from that knowledge rather than override it.
A better test for any river-linking plan
The old test for river interlinking asks whether water can be moved.
The new test must ask whether moving that water leaves the monsoon, the delta and downstream communities safer than before.
That test should be public, rigorous and transboundary. It should not rely on narrow project reports that count benefits in command area and megawatts while treating the atmosphere and sediment as externalities.
A credible assessment would begin with five principles.
Account for feedback, not just flow.
Hydrological planning must include land-atmosphere modelling. If irrigation changes evaporation and surface temperature, those effects belong inside the project boundary.
Treat sediment as infrastructure.
Sediment builds deltas, supports mangroves and shapes channels. A project that traps or diverts sediment should measure that loss as seriously as it measures storage.
Model the coast with the river.
The Ganga-Brahmaputra delta cannot be understood from upstream discharge alone. Sea-level rise, subsidence, tides, storm surge and salinity must be part of the assessment.
Consult Bangladesh before decisions harden.
Shared rivers require shared science. Formal consultation should happen before implementation choices become irreversible.
Compare with demand-side and local water options.
River interlinking should not be evaluated against doing nothing. It should be compared with watershed restoration, groundwater governance, crop policy reform and decentralised storage.
These principles would not automatically kill every inter-basin transfer. They would do something more useful. They would reveal which proposals survive real scrutiny and which depend on leaving out the most dangerous consequences.
A river-linking plan that cannot account for rainfall feedback, sediment loss and downstream harm is not a climate adaptation plan. It is a climate risk with concrete lining.
The political challenge is that large infrastructure creates its own momentum. Once contracts, alignments and institutional interests form, evidence can become an inconvenience. That is why peer-reviewed science matters before projects become facts on the ground.
The studies now before us do not ask India to abandon water security. They ask India to define water security more honestly. Security is not achieved if dry basins receive canals but lose rainfall. It is not achieved if the Sundarbans lose sediment and Bangladesh faces deeper salinity intrusion. It is not achieved if flood control in one region increases vulnerability in another.
True water security must be measured across the whole system.
The choice is between control and wisdom
River interlinking appeals to an old dream: control the rivers, and the nation controls its future. In an age of climate instability, that dream is understandable. It is also dangerous if control means ignoring the systems that make rivers alive.
The monsoon does not obey canal logic. The Sundarbans cannot be rebuilt with compensation money once sediment supply declines. Bangladesh cannot be treated as a downstream detail in decisions that affect shared rivers. The delta cannot be separated from the basin, and the basin cannot be separated from the sky.
The uncomfortable arithmetic is now clearer than it was decades ago. The studies that might once have prevented a bad decision now describe, with sharper numbers, the disaster critics warned about from the beginning.
There is still room for a wiser path. It begins by admitting that rivers are not pipes, sediment is not waste, and rainfall is not guaranteed. It requires India to bring atmospheric science, delta science and transboundary ethics into the centre of water planning.
A country does not become climate-resilient by moving risk out of sight. It becomes resilient by seeing the whole system before it acts.
