J.Konstapel,Leiden,19-8-2026.

Aanleiding

Artikel FD van vandaag: “Een woestijn in het hart van Europa: vruchtbare Hongaarse landbouwgrond verandert in dorre vlakte” (van: David Kabel).

Hungary Is Not Short of Water

There is a photograph of a man standing in a pit he dug on his own land. He digs it every summer, in the same place, to find out how far the water has fallen. Last year he reached it at six and a half metres. This year he had to go down to eight.

He is not a hydrologist. He is a mechanical engineer who lives near Kiskunmajsa, on the sandy ridge between the Danube and the Tisza. He started digging because he found dead rabbits on his land.

That pit is the most useful instrument in Hungarian water management. It measures the one thing that matters and nobody was measuring.

The diagnosis everyone agrees on is the wrong one

The Hungarian plain is drying. That much is not in dispute. Central Hungary is warming about half again as fast as the rest of Europe. The Carpathians hold the heat in. The rain is falling less often and the heat and the dryness reinforce each other.

From this everyone draws the same conclusion. There is less water. Therefore Hungary has a water shortage. Therefore the situation is beyond repair, or repairable only at enormous cost.

The national hydrological association says the plain is past its tipping point and that coming back will be difficult and expensive. The agricultural sector is described as a dying body kept alive by an artificial lung of subsidy. The new government inherits the problem with a budget deficit of eight per cent and no room to spend.

That is a coherent picture and it is wrong in one specific place. It confuses three different things and treats them as one.

The first is how much water arrives. That is set by the circulation over the continent. Nothing done in Hungary changes it.

The second is how much of what arrives is held. That is set entirely by what is done in Hungary.

The third is whether the held water is returned to the local air or exported down a canal. Also set entirely by what is done in Hungary.

Only the first of these three is beyond a tipping point. The other two are arithmetic.

The arithmetic

The ridge at risk of turning into desert covers roughly ten thousand square kilometres. That is a million hectares.

In sand, about a fifth of the soil volume fills and empties with the water table. So raising the water table across that whole area by one metre requires about two billion cubic metres of water. Two cubic kilometres.

Two cubic kilometres is roughly the volume of Lake Balaton.

At which point the number sounds impossible, and the case for despair looks proven. It is not proven. It depends on what you compare it to.

The Tisza carries about twenty-five cubic kilometres of water past that plain every year. The Danube carries about seventy-four.

So the amount needed to lift the aquifer a full metre is about eight per cent of one year’s Tisza. It is under three per cent of one year’s Danube.

Hungary is not short of water. Hungary is short of water that stays.

What was actually done to the plain

This is not an accident of climate and it did not start in 2022.

From 1846 the Tisza was regulated. The meanders were cut through, the river was shortened by about a third of its length, and the floodplain that used to spread across the Alföld every spring was closed off behind dikes.

It worked. Marsh became arable land. That project is the reason the plain is grain country at all, and the reason the Hungarians who settled a wetland ended up farming a prairie.

It also converted a landscape that held water into a landscape that conveys it. The canals that followed were built to move water off the land, into the rivers, and out of the country.

That system still exists and still works perfectly. In a drought it is doing exactly what it was designed to do in a flood.

When a group of volunteers put wooden planks across a channel last autumn and let the water spread onto the meadows, they were closing something that was deliberately opened in the nineteenth century. They did it without a budget, and the fields they flooded are green while the neighbouring one is not.

You can see the boundary in the photographs. One landowner declined to take part. His grass stops at the property line.

The lever nobody is using

There is a second reason this matters, and it is larger than the water balance.

It takes energy to evaporate water. About two and a half million joules per kilogram. Work that through and one extra millimetre of evaporation per day, over one square metre, carries away about twenty-eight watts.

The global warming forcing that all of climate policy addresses is about two point seven watts per square metre.

So one extra millimetre of local evaporation per day moves roughly ten times the global figure, locally.

Across the million hectares of the ridge, that comes to about two hundred and eighty gigawatts. The Paks nuclear station, which supplies forty per cent of Hungarian electricity, produces about two.

This is not a claim that landscape water beats climate change. It is a statement about where the largest available lever in Hungary actually is. Energy that leaves the surface by evaporating water does not heat the air above it. Energy leaving dry ground does. Part of that fifty-per-cent-faster warming is being manufactured on the ground, and the part that is manufactured on the ground can be switched off.

The Danube fell low enough this summer that Paks had to throttle back and households were asked to use less electricity. The cooling capacity of the plain is a power system question, not only a farming one.

The two sunflower fields

Near the Serbian border there are two adjacent fields. Same soil, same rain, same summer.

On one the sunflowers hang their heads and the leaves are dead against the stem. On the other the stems are green and the flowers are up.

The farmer with the green field lost fifteen per cent of his harvest. Everyone around him lost more than half. He has been farming this way for twelve years and until this summer his neighbours thought he was strange. His fields look neglected. They are full of weeds and straw. He leaves ground fallow and lets pigs and cattle work it.

Here is what that is worth, in millimetres.

The top thirty centimetres of a hectare weighs about four thousand tonnes. One percentage point of organic matter in it is therefore forty tonnes. Organic matter holds two to four times its own weight in water. That is somewhere between eight and seventeen millimetres of extra water available to the crop, for each percentage point built.

Add the straw and weed cover, which stops the sun from taking water out of bare ground before a root reaches it. Over a season that is worth a few tens of millimetres more.

Twelve years of this buys something in the region of fifty millimetres.

Fifty millimetres does not sound like a harvest. It is one, because yield is not a smooth function of water. It is a threshold. The last fifty millimetres are the difference between a seed that completes its cycle in July and one that does not. His neighbours were not generally worse off. They were fifty millimetres short at the one moment it counted.

That experiment has already been run, at no public cost, with a control field on the other side of the fence. It is better evidence than most agricultural trials produce.

And the limit of it

Fifty is not two hundred.

The structural shortfall on that ridge — what the sun would evaporate against what the sky delivers — is about two hundred millimetres a year. Field practice closes the gap within a season. It does not refill an aquifer.

The pit at Kiskunmajsa will keep getting deeper no matter how much straw the region leaves on its fields.

This has to be said plainly, because the enthusiasm now building around regenerative farming will otherwise do the damage that enthusiasm usually does. Farmers will adopt it, it will work, the aquifer will keep falling, and in ten years the method will be declared a failure for not solving a problem it was never at the right depth to solve.

Field practice and landscape retention are two different depths of the same net. Neither is sufficient. Together they are a programme.

Hold back an extra fifty millimetres a year across the ridge and you capture half a cubic kilometre annually. Against the two cubic kilometres a metre costs, that is four years per metre. Not a rescue. A slope, measurable every year, running for a decade.

Fifty millimetres a year is two per cent of what the Tisza carries past.

The money argument is already settled

Six billion euro of damage, spread over a million hectares, is six thousand euro per hectare per bad year.

That is the ceiling. Spend less than that per hectare and restoration is cheaper than the compensation currently being paid.

Hungarian output is on the order of two hundred billion. Six billion is around three per cent of it. The deficit that is said to make investment impossible is eight per cent. A substantial share of that deficit is the drought being paid for rather than addressed.

Compensation buys the same bad year again, indefinitely, and leaves nothing behind. Retention is a capital cost with a rising asset under it.

The cheapest item on the entire ledger is a plank across a canal. The volunteers found that without a ministry.

What is actually hard

Not the engineering. Not the money. The boundary in the grass.

Retention is a property of a catchment, not of a parcel. It cannot be bought one field at a time, because the man who refuses is drying out the water table under the man who agrees. There is no technical answer to that. There is a legal one, and it does not exist yet: an instrument that permits controlled retention across an existing canal network, and a rule for who carries the cost of a wet field in a wet year.

That is the thing to put in front of a government. Everything else in this article can be started by people who already control the land they stand on.

One dial

If this becomes a programme it will need a single number, or every institution will measure its own thing and nothing will be settled.

Four readings, all of them already collected in Hungary. The annual change in water table depth across the monitoring wells. The share of rainfall that leaves the block down a gauged canal. The share of summer energy leaving as evaporation rather than as heat, taken from thermal satellite data. And how long the ground stays green through July and August — duration, not peak.

Three of those four can be computed backwards to the 1980s before anyone lays a plank. The baseline costs nothing.

What is being claimed, and how it can fail

The commitments are priced in the accompanying paper, with dates. The short version.

Close the canals in a block and the water table slope should turn positive within three growing seasons. I put that at sixty-five per cent, against roughly twenty for a block left alone.

The gap in summer evaporation between treated and untreated blocks should exceed a tenth of available energy in at least one summer between 2027 and 2029. Sixty per cent.

The yield gap between established regenerative farms and their neighbours in the next severe drought should be at least twenty percentage points. This summer it was around thirty-five. Fifty-five per cent.

Fifty millimetres a year of retention should produce at least twenty centimetres a year of water table rise. Fifty per cent — and if retention is achieved and the table does not move, the sand holds less than assumed and every timescale here stretches.

And the weakest one. Treated blocks should run half a degree to a degree and a half cooler in summer than untreated ones. Forty-five per cent. This is the claim most likely to fail, because a single block is probably too small for the signal to separate from the wind. If it fails at block scale it should be retried at county scale before the whole lever is thrown away.

That last paragraph is there on purpose. A position that cannot lose is not a position.

The three instruments nobody commissioned

A man dug a pit and found the number.

A farmer let the weeds grow and kept his harvest.

A group of volunteers put planks in a canal and turned a meadow green.

None of that was funded, planned or published. All three are better measurements than the debate currently running above them. The work now is to read them at the right depth, put a dial on them, and hand it to people who can do it across ten thousand square kilometres instead of ten.

---

Reading list

1. David Kabel, “Een woestijn in het hart van Europa,” Het Financieele Dagblad, 19 August 2026. Why read? Everything observed in this piece comes from here: the pit, the two sunflower fields, the planks, the refusing landowner, the reactor throttling back. Reading advice: read it as a set of measurements that happen to be written as a story.

2. Imre Pálfai, collected work on the water balance of the Danube–Tisza Interfluve and the Hungarian drought index. Why read? The groundwater deficit on that ridge was documented from the 1980s. It was known for forty years before it became news. Reading advice: this is where to get the real Hungarian numbers for soil storage and shortfall, in place of the round figures used above.

3. László Somlyódy (ed.), Magyarország vízgazdálkodása, Hungarian Academy of Sciences, 2011. Why read? The standing national water strategy. It contains the institutional map — who controls the canals, who controls the wells, who compensates whom — which is where the boundary-in-the-grass problem actually lives.

4. Dénes Lóczy (ed.), Landscapes and Landforms of Hungary, Springer, 2015. Why read? For the Tisza regulation as a physical event: the cut meanders, the shortened river, the closed floodplain. Reading advice: read the Alföld chapters and notice how young the present landscape is.

5. Wilhelm Ripl, “Water: the bloodstream of the biosphere,” Philosophical Transactions of the Royal Society B, 2003. Why read? The best existing statement of the energy argument made here, by someone who worked on real catchments rather than models. Landscapes as structures that dissipate energy, with water as the medium and export as the measure of damage. Reading advice: if you read one item on this list, read this.

6. Michal Kravčík et al., Water for the Recovery of the Climate — A New Water Paradigm, 2007. Why read? Central European, written by people who built the retention works before they wrote the theory. Reading advice: the arithmetic is looser than what is set out above, but the programme is essentially theirs and it has been running for twenty years next door.

7. Sonia Seneviratne et al., “Investigating soil moisture–climate interactions in a changing climate,” Earth-Science Reviews, 2010. Why read? The standard treatment of how a landscape splits incoming energy between evaporating water and heating air, depending on how wet it is. This is the twenty-eight watts, done properly.

8. Diego Miralles et al., “Mega-heatwave temperatures due to combined soil desiccation and atmospheric heat accumulation,” Nature Geoscience, 2014. Why read? Shows a dry surface raising its own air temperature over successive days. It is the clearest published account of drought that intensifies itself.

9. Adriaan Teuling et al., “Contrasting response of European forest and grassland energy exchange to heatwaves,” Nature Geoscience, 2010. Why read? Two adjacent land covers, same weather, opposite energy behaviour. The continental version of the two sunflower fields.

10. Andrea Basche and Marcia DeLonge, “Comparing infiltration rates in soils managed with conventional and alternative farming methods,” PLOS ONE, 2019. Why read? Puts numbers on what cover and residue do to water entering the soil, across many sites. Reading advice: use it to check whether fifty millimetres is generous or conservative on sand.

11. IPCC Sixth Assessment Report, Working Group I, Chapter 11 and the regional Atlas for Central and Eastern Europe. Why read? The source behind the fifty-per-cent-faster warming figure, and the projected drying of the Carpathian Basin.

12. Edward Cook et al., “Old World megadroughts and pluvials during the Common Era,” Science Advances, 2015. Why read? The European tree-ring drought atlas. The Carpathian Basin has had multi-decade dry spells before anyone burned coal. Reading advice: this is the argument against panic, and for measuring persistence instead of reacting to summers.