River-linking plan: India’s river-linking debate has acquired fresh momentum. At the Southern Zonal Council meeting in Mahabalipuram on August 20, Union Home Minister Amit Shah revived the idea of connecting major rivers from the Brahmaputra towards the Cauvery and Godavari. Such a plan will inevitably be judged on engineering feasibility, environmental costs and interstate water claims. There is, however, an equally important issue of governance. If water is to be transferred across basins, the states and agencies involved must first agree on the information that will determine when a transfer is possible, how much water is available and what consequences may follow.
This is particularly important because the National Perspective Plan does not envisage one continuous canal running from the Brahmaputra to southern India. It comprises 30 proposed river links, 14 under the Himalayan component and 16 under the Peninsular component, with several reservoirs and basin interfaces along the way. Ken-Betwa remains the only priority link to have entered implementation. Any larger network will therefore take shape incrementally. Each new connection will add another set of operating decisions involving reservoir levels, downstream requirements, flood conditions and competing claims among states. The quality of the information behind those decisions will matter almost as much as the physical infrastructure.
READ | Decline of Nayar river reveals need for Namami Gange course correction
A surplus must be established in real time
The National Water Development Agency’s studies seek to identify water that can be transferred after meeting the reasonable future needs of the basin and states concerned. That is a necessary basis for long-term planning. It is less straightforward as an operating rule.
Conditions within a basin change through the season. Heavy rainfall in one catchment can coincide with limited reservoir capacity elsewhere. Downstream irrigation demand may be high. Environmental flows have to be protected. Sediment can constrain infrastructure. The rainfall expected over the following week may alter the calculation again.
The relevant operational question is therefore whether a given volume of water can safely be transferred at a particular time, after accounting for conditions in the donor basin and the basin receiving it.
River discharge alone cannot answer that. Operators would need current information on rainfall, catchment conditions, reservoir storage, downstream demand, flood risk and the state of the receiving basin. They would also need to know how reliable the underlying forecasts are.
Once rivers are physically connected, the quality of these decisions becomes part of the infrastructure.
READ | Legal personhood to sacred rivers: The ruling gave symbolism, not solutions
India already has much of the information
India is not starting from scratch. The Central Water Commission has developed rainfall-runoff models that provide seven-day advisory flood forecasts, alongside shorter-range forecasts. C-Flood, the unified inundation forecasting platform, provides two-day forecasts down to village level. Its initial coverage includes the Godavari, Mahanadi and Tapi basins, using two-dimensional hydrodynamic modelling together with satellite data and ground observations.
The Brahmaputra Board is expanding basin master planning using GIS, remote sensing and hydrological modelling. Its current work also covers flood and erosion management and sediment studies, including work on the Siang and Dibang. NRSC and ISRO provide satellite observation, flood mapping and geospatial support for disaster management.
Much of the technical capability therefore exists. The institutional difficulty is that these systems were created by different agencies for different purposes. Reservoir operators, state irrigation departments, weather forecasters, disaster-management authorities and central water agencies do not necessarily work from one continuously updated operating picture.
That fragmentation matters during a flood. A hydrological event can interrupt transport, damage infrastructure, isolate settlements and alter reservoir operations at the same time. Forecasting where water will go is only one part of the decision. Authorities also have to establish what is exposed, what happens downstream and how an intervention in one basin changes risk elsewhere.
READ | River-linking a great idea, but must be done without impacting people, nature
Build a river-intelligence layer
One possible model comes from the proposed Saptarishi Framework, originally conceived as a seven-layer digital architecture for the built environment. Its value for river management lies less in its nomenclature than in its division of institutional functions and its emphasis on interoperability.
Adapted to rivers, one layer could map physical assets such as dams, barrages and embankments. Another could cover land, settlements and floodplains. Network information could connect rivers, canals, roads and utilities, while an environmental layer brings together rainfall, hydrology, groundwater, inundation, erosion and sediment data. Other layers could relate these inputs to economic exposure, administrative authority and emergency response.
This does not require an enormous central database into which every agency surrenders its records. A federated system would allow the competent institution to remain custodian of its data while common standards, identifiers and interfaces let authorised users assemble the information needed for a particular decision.
That distinction is important in water governance. States are unlikely to trust a system whose assumptions they cannot inspect or whose data provenance is unclear. Interoperability is useful only when the information remains auditable.
Test river links before operating them
The strongest case for such a system is that India need not wait for new canals to be completed before testing how they might operate.
Selected links could first run in a digital shadow environment alongside existing river operations for several monsoon seasons. The exercise would alter neither water entitlements nor reservoir operations. It would record what the model would have recommended under actual conditions and compare those recommendations with what subsequently happened.
Historical flood events could also be replayed. The simulations could estimate how much transferable water was genuinely available, whether a transfer would have reduced a flood peak, what environmental flows would have remained and what happened to risk in the receiving basin. Pumping requirements and infrastructure constraints could be incorporated where relevant.
Forecast error should be part of the exercise. A transfer that appears sensible under one rainfall forecast may look very different if the forecast is wrong.
The value of shadow operation lies precisely in its capacity to expose such assumptions before they become operating practice. Some simulations may strengthen the case for particular links. Others may show that the benefit is modest or that risk is shifted downstream or into another basin.
A decision system that can produce evidence against a proposed intervention is more credible than one designed to validate it.
River interlinking is also a federal problem
Physical integration would create more interfaces between states that already dispute water allocations. The technical questions soon become institutional ones. Which measurement establishes that transferable water is available? Which forecast is used when models differ? Who determines the operating threshold? How are environmental requirements incorporated? Can each state independently verify the information on which a release is based?
These questions require agreed rules and traceable data as much as hydraulic expertise.
That makes a common river-intelligence system relevant to cooperative federalism. Zonal Councils already exist to give the Union and states a forum for resolving shared issues. At the August 20 meeting, the southern states agreed to pursue pending water disputes through dialogue involving the states and Union institutions. A shared operating picture would not eliminate competing interests, but it could make the evidence behind competing claims easier to test.
River interlinking will continue to be judged on hydrology, ecological costs, engineering feasibility, economics, displacement and interstate rights. Digital modelling cannot settle those questions in advance.
Its purpose should be more demanding: to show decision-makers when a transfer makes sense, when it does not, and when the evidence is too uncertain to justify one.
Before India commits itself to operating an interconnected river system, the institutions responsible for those rivers should be able to see, test and argue from the same evidence.
Apurva Pathak is an architect and the author of the Saptarishi Framework, a proposed seven-layer Digital Public Infrastructure architecture for the built environment.