Part I — Situation overview

On Saturday morning, 15 August 2026, the controlled sinking of two barges, each 80 metres long, began on the Paks stretch of the Danube. On both banks the bottom sill is also being built — the low stone dam laid into the river bed which raises the water over a short stretch — for which, according to ATV’s report, 145 thousand cubic metres of stone are being used. At the on-site press conference it was said that some 130 people are working on the intervention, fifty are handling the administration, more than a hundred lorries are transporting the stone, and five thousand tonnes of stone arrived that day. Helicopters of the defence forces are lifting two-tonne concrete blocks into the bed. On Saturday morning the prime minister indicated that without intervention the water level would sink below the −132 centimetre threshold by dawn on Monday, which would force the shutdown of one of the two operating turbines. Below −140 centimetres the complete disconnection of the plant would also come into question. The first result is modest but points in the right direction: above and below the barges placed crosswise a level difference of 7–8 centimetres was measured by Saturday morning, while at the cold water inlet of the plant this meant a rise of only one centimetre — for the nominal capacity 25 centimetres would be needed compared with the present level.

On the same day crisis management crossed into another terrain. Foreign minister Anita Orbán announced that following talks with Tomáš Taraba, the Slovak deputy prime minister, a bilateral expert working group had been set up, Hungarian and Slovak water management specialists had sat down at the negotiating table, had agreed on continuous data exchange, and that as a result the necessary discharge had been secured at Nagybajcs. The Hungarian side asked in an official letter that an even discharge be maintained as far as the available water reserves allow.

This second development is the more important one — and in MIAK’s reading also the most delicate. The discharge of the Danube is not a Hungarian resource: from Austria through Slovakia to Hungary it is a single, shared stock whose momentary state is shaped by upstream decisions. Attila Aszódi, nuclear energy expert and professor at the Budapest University of Technology and Economics, recalled in 24.hu’s podcast that already in 2018 he had indicated that the expected water level and temperature of the Danube might make the river unsuitable for the undisturbed supply of the plant once the new blocks entered service, and had urged the redesign of the cooling system — the answer at the time was that the licensing had been concluded and the system would not be touched. The present situation is therefore not an unforeseeable natural disaster but the maturing of a known, documented and set-aside risk. A solution resting on telephone calls and good offices is worth exactly as much as the other party happens to be willing to give.

Part II — Foundations in the literature

The interpretative frame of the topic is available from two authors. In her volume Governing the Commons, Elinor Ostrom, the American political economist awarded the Nobel Memorial Prize in Economics in 2009 for her research on common-pool resources, identified eight design principles which distinguish lastingly functioning common resource systems from those that fail: clear boundaries, rules fitted to local conditions, mutual monitoring, graduated sanctions, quick and cheap conflict resolution, and — in larger systems — nested levels of decision-making. The Danube basin is precisely such a system, and the present Hungarian–Slovak agreement contains one of these elements: data exchange. Nicholas Stern, the British economist and author of the 2006 climate economics report prepared for the British Treasury, says the most important thing about the cost of postponing adaptation: adaptation is the only answer to those effects which will occur in any case in the coming decades, and large infrastructural decisions require foresight and planning. The present stone dumping, improvised within weeks, is the opposite of this. The detailed treatment of the literature — author by author, with quotations — can be found in the 6.4 Literature in detail section.

Part III — MIAK’s concrete proposal

MIAK proposes three measurable measures which replace the acute intervention with a lasting institutional frame.

3.1 A Hungarian–Slovak–Austrian discharge and sediment data sheet (within 90 days)

The present bilateral data exchange is an expert agreement of which neither the content, nor the measurement points, nor the frequency of updating are public. MIAK proposes that the National Directorate for Water Management and its partner institutions should within 90 days set up a joint, machine-readable data sheet which reports, cross-section by cross-section (at least Nagybajcs, Budapest, Paks, Mohács), the discharge, the water level and the temperature on a daily basis, and the sediment balance quarterly. The source of the data sheet is the Hungarian and the Slovak — later the Austrian — hydrological service, the format is uniform, and the historical time series can be downloaded retroactively. This is the minimum of Ostrom’s first and fourth principles (clear boundaries, mutual monitoring): a common resource cannot be managed without everyone seeing the same number. The infrastructure of the K3 pollution monitoring programme point can be extended to this, and the public disclosure follows the logic of KP3 transparent foreign policy: the commitments we make with our neighbours are enforceable if their fulfilment is verifiable by everyone.

3.2 A low-water minimum discharge commitment within the framework of the Danube River Protection Convention (in the first half of 2027)

Telephone goodwill is not a regime. MIAK proposes that in the first half of 2027 the Hungarian government should initiate, in the International Commission for the Protection of the Danube River (ICPDR) — the intergovernmental body established for the protection of the Danube basin — the negotiation of a low-water cooperation protocol: at water levels below a threshold value, what upstream reservoir operating regime the parties undertake, for how long, with what prior notification, and what happens if someone fails to perform. Ostrom’s sixth principle — quick and cheap conflict resolution — is here a concrete request: the protocol should contain an expert conciliation procedure, not a judicial route. The commitment is not unrealistic: the Slovak side has just shown that it is willing. The difference is that a written commitment lives even when no camera is standing on the river bank. This is the practical application of the KP10 regional resilience building and the KP6 multilateral–bilateral strategy differentiation programme points.

3.3 A cooling water exposure test for every power plant decision (from the next update of the energy strategy)

Attila Aszódi’s 2018 warning is of key importance because it shows the missing procedural element: there was no mandatory step which would have made the climate risk of cooling water access a condition of the investment decision. MIAK proposes that for every investment or lifetime extension decision concerning a water-cooled power plant above 300 megawatts, the submitting party should be obliged to attach a public, quantified water resource forecast up to the end of the planned lifetime of the facility, with several climate scenarios, and to indicate the discharge threshold below which the facility cannot be operated. This supplements the risk map of the K7 energy market shock resilience programme point with a hitherto missing dimension, and can be one of the inputs of the K9 domestic climate economics model. This should be the mandatory annex to any decision in any direction concerning the Paks II expansion — continuation, renegotiation or withdrawal.

The common principle of the three proposals is that they treat the crisis not as an event but as a state. Ostrom’s research is about the fact that in lastingly functioning systems the rules are drawn up in periods of calm and the crisis merely tests them; and Stern’s argument is that foresighted adaptation is cheaper than delayed adaptation. The present stone dumping is a counterexample to both — and precisely for this reason it is the best moment for the rules of the next ten years to be born.

Part IV — Expected effects and risks

Dimension Expected effect Risk
Economy More predictable power plant availability, less forced import in low-water periods; a more realistic cost basis for investment decisions The cooling water exposure test slows licensing and may classify some projects as uneconomic
Society Alongside public data the debate on water restrictions stays fact-based, the space for panic news narrows The data in themselves do not resolve the conflict; the clash of upstream and downstream interests becomes more visible
Public administration Coordination between water management, foreign affairs and energy is put into a procedural order rather than remaining ad hoc A new coordination burden; the negotiation of the international protocol may drag on for years, and in the meantime bilateral improvisation remains
External relations Hungary appears on the river basin as a norm-defending actor, not a complaining one The sovereignty sensitivity of the neighbours: upstream countries may see a binding commitment as a restriction of their room for manoeuvre

The main dilemma of the package of proposals runs between speed and bindingness. Obtaining a legally binding minimum discharge commitment is slow, and if it is too strict the upstream partners will not sign it — while the informal agreement that has just proved itself worked immediately. MIAK’s position is that the two are not alternatives to each other: bilateral rapid assistance remains the acute instrument, and the protocol is the basis that can be invoked at the time of the next drought. The proposal may tip over to the risk side if the monitoring is completed but no commitment is attached to it — then the public data will merely document powerlessness. And the cooling water exposure test works if the setting of the threshold value goes to an independent professional body; if the investor itself determines what the critical discharge is, the test empties out into a formality.

Part V — Measurability and summary

5.1 What is worth following? (proposed KPIs)

MIAK recommends four proposed performance indicators (KPIs) to the attention of the public:

  • Publicity of data: from the first quarter of 2027, a daily updated, machine-readable discharge time series from at least four Hungarian and two Slovak cross-sections, with retroactive download possibility.
  • Coverage of commitments: by the end of 2028 there should be a written commitment on the low-water operating regime with at least two upstream partners, with a defined threshold value and notification deadline.
  • Subsequent impact assessment: the water management, shipping and ecological impact assessment of the present intervention should be publicly available within 90 days of its completion.
  • Planning discipline: from 2028, 100% of decisions on water-cooled power plants should be accompanied by a public cooling water exposure analysis — at present there is no such obligation.

5.2 Summary

The Paks intervention is a professionally defensible emergency solution, and the work of the several hundred people working on site deserves recognition. MIAK’s request of the decision-maker is nevertheless that in the coming weeks the measure of success should not be the quantity of stone but whether an institution is born out of the crisis: a joint data sheet, a written low-water commitment and a mandatory cooling water exposure test. None of these is costly, none of them is spectacular, and all three will have an effect even when the river happens to be high.

Two MIAK foundational values are directly in play here. Data-drivenness: one cannot negotiate responsibly about a shared resource if the parties state different numbers — joint, public measurement is not a technical detail but the precondition of the negotiation. And transparency: the story of the 2018 expert warning that was set aside shows precisely what happens if a risk analysis remains an internal matter of the decision-making process. If the cooling water forecast had been a public annex, today’s debate would not be about who warned and who was waved away.

Part VI — Reasoning and further sources

6.1 The press framing by spectrum

The liberal-left band carried both the details of implementation and the accompanying political noise. Telex concentrated on the technical content of the press conference — the three construction phases of the bottom sill, the securing of the barges, the order of turbine reconnection — and separately highlighted that the prime minister had invited the leaders of the parliamentary parties to Paks, as well as the stone offers of large companies linked to the NER. 444.hu put the most important measurement fact into its headline: the barges caused a detectable rise already by being placed crosswise, before their sinking — that is, the paper presented the result in its own order of magnitude, without magnification. Népszava carried the political reactions (headline-level reference only).

The public affairs band emphasised the drama of the situation: ATV framed the work as a “heroic struggle” and placed the rise of the water level at the centre. 24.hu, by contrast, chose the frame that looks the furthest ahead: the podcast made with Attila Aszódi was not about the present intervention but about the expert warnings set aside in 2018 and about the conditions of the Paks II expansion — this is the only one among today’s materials that shows the crisis together with its antecedent.

The business band splits in two in the case of Portfolio. The news about the water management cooperation conveys the diplomatic result on the basis of the foreign minister’s announcement, while the 13 August episode of the paper’s Alapvetés podcast with the environmental engineer Zoltán Goda searches for the structural cause — according to the title, the upstream works hold back not primarily the water but the sediment, which brings with it the long-term deepening of the river bed. This frame explains why a single sill does not solve the problem.

The pro-government and conservative band worked, strikingly, not with criticism of the intervention but with professional confirmation. Mandiner carried Attila Aszódi’s article, according to which the bottom sill is an “urgent and workable solution”, and set out in detail why a sill built at the mouth of the plant’s cold water channel gives the greatest rise in level from the least stone. In MIAK’s reading this is the most valuable element of today’s press picture: the professional soundness of a government measure was confirmed by the band opposed to the government. Magyar Nemzet was present with a presentation of the helicopter work of the defence forces (the article was not publicly downloadable).

6.2 Facts and data

Data Value Source
Water level expected by dawn on Monday without intervention below −132 cm (shutdown of one turbine), possibly below −140 cm (risk of complete shutdown) prime ministerial announcement, 15 August 2026
Level difference measured above and below the barges 4 cm (first day) → 7–8 cm (morning of 15 August) Gyula Kovács, water management technical director / 444.hu
Rise in level at the cold water inlet +1 cm 444.hu, 15 August 2026
Further rise in level needed for nominal capacity +25 cm at the inlet Károly Vigh, operations director / Telex, 444.hu
Stone used for the bottom sill 145 thousand cubic metres ATV, 16 August 2026
Stone transported to the site in one day 5,000 tonnes prime ministerial statement, 15 August 2026
Reserve of the water level of 3 August 2026 a few cm above the operability limit of the cooling water pump of the last turbine; 10 cm at the pumps of the safety cooling water system Attila Aszódi / Mandiner, Portfolio
Reserve attainable by converting the pumps at most 50 cm, with several years of work Attila Aszódi / Mandiner
Minimum Danube discharge measured so far 720 m³/s; the sill would secure operation down to 600 m³/s Attila Aszódi / Mandiner
Difference between low and high water levels in the Paks cross-section more than 9 metres Attila Aszódi / Mandiner

The series of numbers in itself tells the essence of the story. Between the order of magnitude of the intervention (a few centimetres) and the order of magnitude of the problem (more than nine metres of natural fluctuation, a 25-centimetre shortfall for nominal operation) the difference is so great that the bottom sill can be interpreted not as a solution but as a lifting of the threshold: it gives exactly as much as is needed to avoid a shutdown, and nothing more. This is not a criticism of the technical solution — according to Attila Aszódi’s analysis it is precisely the proportionate, quickly implementable answer — but the basis of the planning conclusion: whoever operates a baseload power plant with such a narrow reserve has to increase the reserve itself, not the emergency toolkit.

6.3 Policy dimensions

  • Environment and climate (programme points) — the cooling water risk as a hitherto missing dimension in the risk map of K7 energy market shock resilience; the monitoring infrastructure is an extension of K3; the long-term water resource path is an input to the K9 model.
  • Foreign policy (programme points) — the bilateral water management channel is the bilateral branch of KP6, and the Danube River Protection Convention the multilateral one; the public data sheet is the application of the transparency logic of KP3 to a technical field.
  • Transport and infrastructure (background material) — the bed sill affects the cross-section of the international waterway, so the downstream shipping interest is a mandatory element of the impact assessment.
  • Economy (background material) — the loss of power plant output can be replaced by imports and gas power plant generation, which directly increases exposure in energy supply security.

6.4 Literature in detail

6.4.1 Elinor Ostrom: Governing the Commons — The Evolution of Institutions for Collective Action

The starting point of Ostrom’s work is that neither nationalisation nor privatisation on its own manages common-pool resources well — the lasting solution is a system of rules developed and maintained by the users. She identified eight design principles which regularly appear in long-functioning systems, and of which the last relates expressly to large, complex systems:

“For common-pool resources that are parts of larger systems: 8. Nested enterprises. Appropriation, provision, monitoring, enforcement, conflict resolution, and governance activities are organized in multiple layers of nested enterprises.”

Ostrom adds that monitoring and graduated sanctions are not instruments imposed by an external authority: in functioning systems the users themselves watch each other, and a breach of the rules is followed by mild responses adjusted to the gravity of the infringement. In the case of the Danube this means that enforcement does not mean an EU infringement procedure but the mutual and continuous monitoring of the parties by each other — precisely what a joint, public discharge data sheet makes possible. In Ostrom’s sequence the present Hungarian–Slovak agreement is the first step: the clarification of the boundaries of the system and of the data. What is missing is the second and the sixth element — the rule tied to local conditions on the low-water operating regime, and the quick, cheap conciliation forum that can be turned to in case of a dispute.

📖 Source: Elinor Ostrom: Governing the Commons — The Evolution of Institutions for Collective Action

6.4.2 Nicholas Stern: The Economics of Climate Change — The Stern Review

Stern’s report is to this day the most cited summary of the economic management of climate change, and it is its chapter on adaptation that speaks directly to the present situation. Two of its statements are important. One is that adaptation is not a matter of choice:

“Adaptation is the only available response to those impacts that will occur over the next several decades before mitigation measures can have an effect.”

The other is that part of adaptation happens by itself, but large infrastructural decisions do not belong here: those, in Stern’s words, “require greater foresight and planning”, and delay steeply raises the cost and increases the residual damage. The Paks case is a textbook illustration of this sentence. The cooling system of a power plant belongs among the infrastructural decisions requiring the greatest foresight, because it fixes for decades the water conditions under which the facility can be operated. When the redesign of the cooling system was raised in 2018, that was the moment of Stern’s “early action”; and the 2026 stone dumping is the price of the delay of which the report speaks — not a catastrophe, but orders of magnitude more expensive and more uncertain than the step taken in time would have been.

📖 Source: Nicholas Stern: The Economics of Climate Change — The Stern Review

6.5 International comparison

The cooling water risk is not a Hungarian speciality, and the region is providing the parallels right now. In mid-August Romania shut down the Cernavodă nuclear power plant because of the low water level of the Danube, that is, the same hydrological event led to two different decisions in two countries: one shut down, the other intervened in the river bed. In France the output of the nuclear power plants along the Rhône and the Garonne has for years been regularly restricted during summer heatwaves — there the temperature limit of the cooling water is the bottleneck, not the water level — and the regulation expressly handles the possibility of a temporary derogation, with public conditions.

Two systems deserve attention as institutional models of common resource management. The international commission established for the protection of the Rhine (ICPR) has for decades operated a joint, public measuring network and alert chain, which the Danube region took as a model precisely after the pollution accidents. And the American water-sharing system of the Colorado River provides the other lesson: there the parties have for decades worked with fixed, quantified low-water allocation rules, which has not eliminated the conflict but has made it predictable — the disputes are about how the rule is applied, not about whether there is a rule at all. The Hungarian aim in proposal 3.2 is exactly this: not the elimination of the disagreement but the fixing of the frame within which it runs its course.

Environment and climate

  • K3 — Pollution monitoring
  • K7 — Energy market shock resilience
  • K9 — Domestic integrated climate economics model
  • K2 — Plan of the energy transition

Foreign policy

  • KP3 — Transparent foreign policy
  • KP6 — Multilateral–bilateral strategy differentiation
  • KP10 — Regional resilience building
  • KP7 — Foreign policy crisis management protocol

Proposed new programme point: A cooling water exposure test for large power plant investment decisions — for the Environment and climate area.

6.7 List of sources

Press sources (MIAK press monitor, 16 August 2026 — topic 1):

  • [Telex] Orbán Anita a szlovák miniszterelnök-helyettessel egyeztetett arról, hogy tartsák fenn a stabil vízszintet — https://telex.hu/belfold/2026/08/15/orban-anita-szlovakia-vizugy-egyeztetes
  • [Portfolio] Fontos bejelentést tett Orbán Anita: magyar-szlovák összefogás indult a vízügyi válság kezelésére — https://www.portfolio.hu/gazdasag/20260815/fontos-bejelentest-tett-orban-anita-magyar-szlovak-osszefogas-indult-a-vizugyi-valsag-kezelesere-856514
  • [Telex] Megkezdték a bárkák süllyesztését Paksnál, Magyar Péter jó híreket kapott — https://telex.hu/belfold/2026/08/15/paks-sajtotajekoztato-magyar-peter
  • [444.hu] Még el sem süllyesztették a bárkákat, máris vízszint-emelkedést okoztak — https://444.hu/2026/08/15/meg-el-sem-sullyesztettek-a-barkakat-maris-vizszint-emelkedest-okoztak
  • [Portfolio] Az osztrákok nem a vizünket tartják vissza, hanem a hordalékunkat — https://www.portfolio.hu/gazdasag/20260813/az-osztrakok-nem-a-vizunket-tartjak-vissza-hanem-a-hordalekunkat-856044
  • [Mandiner] „Sürgős és működőképes megoldás": támogatja a paksi fenékküszöböt Aszódi Attila — https://mandiner.hu/kozelet/2026/08/surgos-es-mukodokepes-megoldas-tamogatja-a-paksi-fenekkuszobot-aszodi-attila
  • [ATV] Heroikus küzdelem Paksnál: már emelkedőben a Duna vízszintje — https://www.atv.hu/belfold/20260816/paks-duna-vizallas-magyar-peter/
  • [24.hu] „Már 2018-ban figyelmeztettem Süli Jánost" — Aszódi Attila a paksi leállásról — https://24.hu/fn/gazdasag/2026/08/12/podcast-della-aszodi-attila-paksi-atomeromu/
  • [Magyar Nemzet] Így rakják tele betonkockával a Duna medrét Paksnál a honvédség helikopterei — https://magyarnemzet.hu/belfold/2026/08/igy-rakjak-tele-betonkockaval-a-duna-medret-paksnal-a-honvedseg-helikopterei-video (the article was not publicly downloadable)
  • [Népszava] Magyar Péter elégedett a paksi munkával, keményen vissza is szólt Toroczkai Lászlónak — https://nepszava.hu/ (headline-level reference only)

Knowledge base references (professional literature):

  • 📖 Elinor Ostrom: Governing the Commons — The Evolution of Institutions for Collective Action
  • 📖 Nicholas Stern: The Economics of Climate Change — The Stern Review

Note: the local file path of the books does not appear in the visible text of the blog — only the author and the title. The file path is an internal matter of the generation process, not the reader’s.

MIAK internal materials:

  • MIAK policy area: Environment and climate (programme points; programme point ID: K7, K9)
  • MIAK policy area: Foreign policy (programme points; programme point ID: KP10)
  • MIAK policy area: Economy (background material)
  • MIAK press monitor, 16 August 2026 — topic 1, score: 89/100

Supplementary public data sources:

  • National Directorate for Water Management — water level and discharge time series (Nagybajcs, Budapest, Paks, Mohács cross-sections)
  • International Commission for the Protection of the Danube River (ICPDR) — river basin management plans and measurement data series
  • Copernicus European Drought Observatory — drought indices
  • European Union Water Framework Directive (2000/60/EC) — implementation reports

Generation metadata