As Himalayan glaciers retreat and monsoon patterns collapse into unpredictability, the suspension of the world’s most durable water treaty is no longer a bilateral grievance. It is a governance gap opening at the worst possible moment.
In June 2026 Pakistan’s Meteorological Department issued its second glacial lake outburst flood warning of the month, cautioning that sustained high temperatures across Gilgit Baltistan and Khyber Pakhtunkhwa were pushing unstable high altitude lakes toward collapse. That same season, climate scientists tracking the Upper Indus Basin reported something structurally new: the monsoon is shifting north and beginning to overlap with the high altitude snow and glacier melt season that once arrived on its own separate calendar. Rain now falls on top of an already melting snowpack. The result is a flood regime that no longer follows the timelines Pakistan built its disaster response systems around.
This is the physical reality sitting underneath a political decision made fourteen months earlier and eight hundred kilometers to the east. A day after gunmen killed twenty six people at Pahalgam in April 2025, India announced it was placing the Indus Waters Treaty in abeyance, blaming Pakistan for sponsoring the attack. Fifteen months on, the treaty remains frozen, India’s position unmoved through a Court of Arbitration ruling it rejects, a Pakistani campaign at the United Nations Security Council, and an international legal conference in Islamabad. The world’s most cited example of successful transboundary water diplomacy, in force since 1960 through two wars, a Kargil conflict, and repeated militant attacks, is now suspended precisely as the river system it governs becomes less predictable than at any point in the sixty five years the treaty has existed.
That timing is the argument. Climate stress does not make treaty suspension a manageable inconvenience. It makes it a compounding risk, because the very features climate change is degrading, predictable flow data, coordinated flood warning, agreed storage limits, are the features the treaty was built to guarantee.
Stay ahead of the geopolitical week.
MD Briefing delivers expert analysis across five global fronts — the Indo-Pacific, energy, geoeconomics, European security, and the Middle East — every Monday morning. Free.
A river system running on borrowed stability
The Indus depends on glacier and snowmelt for roughly forty percent of its total flow, and during the most extreme flood events that share can exceed seventy percent, according to hydrological modeling of the Upper Indus Basin. Pakistan holds the largest concentration of glacial ice outside the polar regions, more than thirteen thousand glaciers feeding a river system that around two hundred twenty million people rely on for drinking water, irrigation, and hydropower. Research on flood risk at Tarbela Dam projects that the frequency of critically high water levels there will nearly triple by the end of the century, with earlier snowmelt pushing flood peaks earlier in the year and giving emergency planners less warning time, not more.
None of this is abstract. Pakistan’s 2022 floods, driven by a combination of record monsoon rainfall and accelerated glacial melt, killed over seventeen hundred people and caused economic losses estimated above thirty billion dollars. Research since has found that predicted flood impacts in the basin could rise by up to five hundred percent under higher warming scenarios. Pakistan has responded with its own institutional efforts, a 2025 Glacier Conservation Strategy, wetland restoration under the Recharge Pakistan programme, and the Living Indus Initiative aimed at rehabilitating basin ecosystems. But provincial authorities in Gilgit Baltistan and Khyber Pakhtunkhwa openly acknowledge they lack the funding and technical capacity to build reliable early warning systems for glacial lake collapse. Domestic adaptation, however well designed, cannot substitute for the one mechanism that was supposed to guarantee coordinated, binational data sharing and flood response across the basin as a whole.
That mechanism is the treaty currently in abeyance.
What the treaty actually did, and what its absence removes
The 1960 agreement, brokered by the World Bank, divided the Indus system’s six rivers between the two countries, allocating the western rivers, the Indus, Jhelum, and Chenab, primarily to Pakistan while permitting India limited hydropower use that could not alter the flows Pakistan depended on. Its real achievement was procedural rather than symbolic: a standing Permanent Indus Commission, mandatory data exchange on river flows and works, and a layered dispute mechanism running from neutral experts to international arbitration. For sixty five years, whatever the state of the broader relationship, that machinery kept operating.
It is not operating now. Pakistan’s Indus Commissioner has stated that he wrote four times to his Indian counterpart about unexplained fluctuations in Chenab flows without receiving a reply. India’s government has been explicit that this is not a temporary pause awaiting technical resolution. Home Minister Amit Shah has said the treaty will “never” be restored, and that India intends to ensure no water is allowed to leave Indian territory for Pakistan. Whatever one concludes about the justification, the practical consequence is that the two governments most responsible for coordinating flood warning, glacial lake monitoring, and drought response across a basin now living through unprecedented hydrological volatility are not speaking to each other about it.
The legal contest India has chosen to ignore
Pakistan has not treated the abeyance as a closed matter. It pursued arbitration through the Permanent Court of Arbitration at The Hague, which ruled in 2025 that the treaty contains no provision for unilateral suspension and reaffirmed its own jurisdiction, and again in May 2026 issued an award on storage limits at Indian hydropower projects that Islamabad has read as a confirmation of its core position, that the treaty imposes binding constraints on India’s water control capacity on the western rivers, enforceable at the design stage rather than through after the fact assurances. India has rejected both rulings as void, maintaining the tribunal itself is illegitimate, a position it has held without deviation since the treaty was suspended.
Pakistan has since moved the dispute into more forums rather than fewer. In January 2026 it convened an Arria Formula meeting at the UN Security Council, where its ambassador argued India’s position amounted to a serious violation of international law. By late April, Foreign Minister Ishaq Dar had written directly to the Security Council’s president warning of grave consequences for the two hundred forty million people dependent on the Indus system, and by June he was hosting an international legal conference in Islamabad specifically to build the case that India’s abeyance has no standing in international law, warning that any diversion of Pakistan’s allocated waters would be treated as an act of war. Whatever the merits of the underlying grievance, the pattern is one of sustained, multi track legal and diplomatic engagement, arbitration, Security Council diplomacy, international convening, pursued without escalating toward the kind of unilateral action that would make the current freeze irreversible.
Taking India’s argument seriously
The strongest version of India’s case is not solely about Pahalgam. New Delhi has separately argued, through formal notices filed in 2023 and 2024, that the treaty was negotiated for a hydrological and technological reality that no longer exists, ignoring six decades of population growth, dam engineering advances, and the same climate volatility Pakistan cites as a reason for restoration. India contends its attempts to open a modernization conversation were repeatedly declined by Islamabad, and separately maintains that a treaty resting on good faith cannot obligate full compliance from one party while the other, in India’s account, tolerates cross border militancy against it.
Both points deserve to be taken on their own terms. A treaty engineered around 1960s assumptions about dam technology and demographic scale may well need updating, and there is no serious analyst who argues the IWT is a perfect or timeless instrument. But the treaty’s own Article XII provides a mechanism for negotiated modification. Abeyance is not modification. It replaces a slow, difficult, but rules based renegotiation process with an indefinite freeze that produces no updated agreement, no new storage rules, no revised data sharing framework, only the absence of the old one. Whatever changes the basin’s hydrology genuinely requires, an institutional vacuum is a strange way to arrive at them, particularly when the vacuum falls hardest on the downstream state carrying most of the physical climate risk.
Three paths from here
The most likely trajectory, on current evidence, is continuation of the present standoff: India holds the treaty in abeyance indefinitely while rejecting arbitral authority, Pakistan continues building its legal and diplomatic case across multiple forums without escalating to unilateral infrastructure moves of its own, and the Permanent Indus Commission remains dormant. This outcome, roughly six in ten by our estimate, is the least dramatic and also the most quietly dangerous, since it normalizes the absence of coordinated flood and drought management during exactly the years the basin’s hydrology is becoming most volatile.
A second path, perhaps one in four, involves a narrow, security linked partial restoration, likely limited to technical data exchange on flood warning ahead of a severe monsoon season, without a broader resumption of the treaty’s institutional architecture. This would be driven less by diplomatic breakthrough than by the political cost, on both sides, of a major flood disaster occurring without any warning coordination in place.
The remaining scenario, and the one that matters most for risk planning, is a shift from diplomatic freeze to physical alteration of flows, additional Indian storage or diversion capacity on the western rivers that goes beyond what the Court of Arbitration’s rulings restrict, met by a Pakistani response that moves beyond legal argument. This is a lower probability outcome, but not a negligible one, and it is the scenario in which a bilateral treaty dispute becomes a regional security crisis between two nuclear armed states.
The indicator worth watching
The clearest near term signal will not come from statements, both governments have made their positions plain, but from conduct around the next major monsoon season. Whether India responds to Pakistan’s outstanding Indus Commissioner correspondence on Chenab flow fluctuations, and whether any flood or glacial lake data is exchanged ahead of peak melt season, will indicate whether the abeyance remains a diplomatic freeze or is hardening into something closer to operational silence during a period of genuine physical danger. A monsoon season that passes with continued silence on data sharing, following a period of heightened GLOF alerts, would mark a meaningful deterioration regardless of what either government says publicly.
What this means for institutional readers
For Western policy planners, the Indus dispute should be read as a slow onset security risk rather than a resolved bilateral matter, warranting the same kind of standing monitoring attention given to other nuclear armed state frictions, not because conflict is imminent, but because the absence of coordination mechanisms narrows the margin for error during climate driven disasters.
For institutions with financial exposure to Pakistan, sovereign lenders, agricultural commodity traders, and infrastructure investors alike, treaty uncertainty is now a variable that belongs in country risk models alongside currency and political risk, particularly given Pakistan’s dependence on Indus irrigated agriculture for export earnings.
For multilateral institutions, the World Bank’s own credibility as the treaty’s guarantor is now part of the story. An instrument the Bank helped broker and has facilitated for sixty five years is being tested by a dispute the Bank has been reluctant to adjudicate directly, and that reluctance itself has consequences for how seriously other transboundary water agreements the Bank has brokered elsewhere will be treated in future disputes.
