River Kaveri in Kodagu. (Image: Rameshng/Wikimedia Commons)
Some rivers carry more than water. The Kaveri carries a century of agreements, tribunals, verdicts and unresolved trust between two states, and every few years, when the monsoon falters, all of it resurfaces at once.
In the last week of July 2026, the Kaveri Water Regulation Committee ordered Karnataka to release 3,500 cusecs of water daily to Tamil Nadu for fifteen days, and the Kaveri Water Management Authority upheld the order despite Karnataka's objection that a severe monsoon shortfall left it unable to comply. Karnataka argued that irrigation in its own command area had not even begun for want of rain. Tamil Nadu called the quantum inadequate, having sought roughly 9.8 tmc ft against an order that would deliver closer to 4.5 tmc ft.
Made using Gemini AI
Within days, farmers and Kannada organisations staged demonstrations near the KRS reservoir, and protests spread to Mandya, Mysuru and Ramanagara with threats of a state-wide bandh, while Tamil Nadu moved the Supreme Court to secure what it considers its legitimate share. A bilateral visit by Tamil Nadu Chief Minister C. Joseph Vijay to Bengaluru, planned to defuse tensions through direct talks with Karnataka's D.K. Shivakumar, was itself deferred after Kannada organisations and farmer groups objected to talks happening under protest.
None of this is new in form. What is worth examining is why a dispute governed by a tribunal award, a Supreme Court verdict, and two purpose built regulatory institutions still detonates almost every year that the monsoon underperforms, and what that recurrence says about the design of India's interstate river governance, rather than simply about the two states' politics.
The Kaveri basin dispute is older than independent India. Its origins lie in 1892 and 1924 agreements between the Madras Presidency and the princely state of Mysore, under which Mysore was permitted to build the Kannambadi, or KRS, dam, subject to conditions protecting the delta region's prior irrigation rights. Those agreements set a pattern that persists structurally to this day: an upper riparian state building storage and diversion capacity, and a lower riparian delta economy dependent on the predictability of releases timed to its cropping calendar.
Modern adjudication began with the constitution of the Kaveri Water Disputes Tribunal in 1990, which, delivered its final award after seventeen years of hearings, allocating the basin's assessed yield of roughly 740 tmc ft among Tamil Nadu, Karnataka, Kerala and Puducherry. Both principal states challenged the award before the Supreme Court, which delivered its verdict on 16 February 2018.
The Court largely upheld the tribunal's allocation but revised the shares, notably granting Karnataka an additional 14.75 tmc ft, of which 4.75 tmc ft was earmarked for Bengaluru's drinking water needs, a city that sits outside the Kaveri basin altogether but depends on it for a large share of its municipal supply. This adjustment matters for the current crisis. It embedded a growing, non-agricultural, politically non-negotiable claim on the river's water into an allocation framework originally designed around irrigation economics.
To operationalise the 2018 verdict, the Centre notified the Kaveri Water Management Authority in 2018, with the Kaveri Water Regulation Committee functioning beneath it as the technical body that recommends month wise release schedules based on rainfall, storage and inflow data. On paper, this two-tier structure was meant to convert an adversarial legal dispute into a technocratic, rule-based administrative process, precisely what such disputes need to escape their cycle of litigation and street protest.
In practice, the Authority has struggled to hold that ground. Its orders are advisory in character until reinforced by the Supreme Court, so every deficit year reopens the same argument through the same forums: Karnataka pleads inability, Tamil Nadu escalates to contempt or fresh writ petitions, and the Court is drawn back into adjudicating operational releases it was never meant to manage.
The Supreme Court itself signalled this frustration in November 2025, when it dismissed as "totally misconceived" a Tamil Nadu application seeking to block the Authority from even discussing Karnataka's Mekedatu detailed project report, while affirming that Karnataka remains bound to release the water allocated to Tamil Nadu and Puducherry regardless. The bench's direction, that operational disputes belong before the Authority, the Regulation Committee or the Central Water Commission, decided expeditiously, is a reasonable division of labour in theory. It has not, so far, stopped either side from treating the Court as the ultimate arbiter of week-to-week case figures, which is a poor use of apex judicial time and a poor substitute for a genuinely rule-bound distress-sharing mechanism.
Running parallel to the annual release conflict is the more structurally consequential fight over Karnataka's proposed Mekedatu balancing reservoir and drinking water project, a roughly 67 tmc ft storage structure with an estimated cost north of ₹5,900 crore, intended primarily to secure Bengaluru's drinking water and generate some hydropower. Karnataka's position, restated by Shivakumar as recently as this crisis, is that Mekedatu is "the only solution" to the recurring dispute, since it would give Karnataka the regulating capacity to smooth releases across dry and wet years rather than negotiating each shortfall from a position of empty reservoirs.
Tamil Nadu has opposed the project since it first surfaced in serious form around 2018, arguing that any new storage structure on the main river, built without its concurrence as the lower riparian state, violates the letter and spirit of the 2007 tribunal award and the 2018 Supreme Court verdict, and would give Karnataka discretionary control over flows that are supposed to be governed strictly by the tribunal's monthly schedule.
This is not merely a technical disagreement; it is a trust deficit. Tamil Nadu's objection assumes that additional storage capacity in upstream hands will, sooner or later, be used to withhold rather than merely regulate water, an assumption Karnataka understandably calls unfair given that, as this year's data show, it has historically over released rather than under released relative to its tribunal obligation.
This is where the 2026 crisis becomes genuinely interesting rather than merely repetitive. Karnataka's own water flow records at the Biligundlu gauging station show that between 2016 and 2026, the state released about 2,746 tmc ft of water to Tamil Nadu, roughly 973 tmc ft more than the 1,772.5 tmc ft the 2018 Supreme Court judgement actually required over that decade. Farmer organisations in Mandya, where 167 of 968 lakes have run completely dry and hundreds more are below a quarter of capacity, argue that this surplus should have been captured through desilted tanks and improved storage infrastructure rather than allowed to flow downstream in wetter years, leaving the state with no cushion when a deficit year like 2026 arrives.
That single data point complicates the standard framing of this dispute as a zero-sum contest that Tamil Nadu perpetually wins and Karnataka perpetually loses, or vice versa. It points instead to a shared institutional failure: the absence of a genuine distress sharing formula that adjusts allocations proportionately to actual rainfall performance across both states, and the absence of investment, on Karnataka's side, in capturing surplus year flows for use in deficit years, which is precisely the storage management problem Mekedatu claims to solve, and which Tamil Nadu's objection, if it prevails indefinitely, leaves permanently unaddressed.
Karnataka's claim that Mekedatu would end the annual crisis, and Tamil Nadu's claim that Mekedatu would enable water withholding, are not straightforwardly reconcilable, but neither is obviously wrong given the state of trust between the two governments. That unresolved tension, not any single season's rainfall, is the real structural driver of the recurring conflict.
Made using Gemini AI
Beneath the legal and political architecture, the dispute has direct livelihood consequences on both sides of the border. In the Kaveri delta, the annual opening of the Mettur dam's shutters on 12 June is the trigger for kuruvai (short-duration) paddy cultivation; insufficient storage in Mettur and poor inflow from Karnataka this year meant most farmers without private pump-set access could not undertake kuruvai sowing at all, pushing the season's crop prospects toward an uncertain October harvest.
In Karnataka's own Kaveri basin districts, drying lakes threaten not just irrigation but drinking water security for both people and livestock heading into the dry months — a concern that has driven farmer leaders to appeal to the state government to reject the CWRC's release directive outright and prioritise domestic supply over inter-state obligations.
Both sides, in other words, are not performing distress for negotiating leverage; the underlying scarcity is real and simultaneous, which is exactly the scenario the current adjudicated allocation framework — built around fixed monthly quotas rather than dynamic, rainfall-indexed distress formulas — is least equipped to handle gracefully.
The current flashpoint carries an added layer of volatility because it coincides with a genuinely new political configuration in Tamil Nadu, where actor-turned-politician Joseph Vijay now occupies the chief minister's office, and in Karnataka, where D.K. Shivakumar holds both the chief ministership and the water resources portfolio he has long used to advance the Mekedatu case.
The CWMA's order arrived, by most accounts, at a politically sensitive moment for both governments. Both leaders face domestic political incentives to be seen as uncompromising defenders of their state's water rights — Shivakumar under pressure from Kannada organisations and Mandya-Mysuru farmer groups warning of an escalating agitation, and Vijay under pressure from a Tamil Nadu political class, including opposition voices, watching closely for any sign of softness on Kaveri, historically one of the most emotionally charged issues in Tamil Nadu politics. The deferred Bengaluru visit is itself a symptom: even the appearance of bilateral goodwill has become politically costly for the upper riparian side while its own farmers are protesting in the streets.
Three structural reforms would do more to de-escalate this cycle than any single season's negotiation. First, a codified, judicially sanctioned distress-sharing formula — proportionate release cuts for both states in years of assessed basin-wide deficit, rather than case-by-case CWRC recommendations contested afresh each time — would remove much of the annual litigation and street-protest cycle by replacing discretion with a pre-agreed rule.
Second, Karnataka's surplus-year over-releases point to underinvestment in intra-state storage and tank rehabilitation that could buffer deficit years without reopening the inter-state allocation question at all; this is a domestic water-governance failure Karnataka can address unilaterally regardless of how Mekedatu is resolved.
Third, resolving Mekedatu will require Tamil Nadu and Karnataka to negotiate not the reservoir's existence but its operating protocol — binding, third-party-monitored release rules that make withholding contractually and technically difficult — since outright rejection has not stopped Karnataka's pursuit of the project through successive governments, and continued deadlock guarantees that the underlying storage-management problem Mekedatu is meant to solve simply recurs, unaddressed, every deficit year.
Made using Gemini AI
The Kaveri river dispute has outlived the agreements, the tribunal, the verdict and the authorities built to settle it, not because any of those instruments were poorly conceived, but because none of them were designed for a basin where rainfall variability, urban drinking water demand and interstate political incentives are all rising at once.
Until the framework is updated to reflect that reality, the people who depend on the Kaveri, farmers in Mandya and the delta alike, will keep bearing the cost of a dispute the institutions built to end it have not yet learned to resolve. The 2026 crisis will not be the last of its kind. It is only the most recent.