Executive Summary

Researchers from WSL, University of Bern, and Eawag have created the first comprehensive database on residual water determinations at hydroelectric power plants. Simulations show: The additional production reduction due to new concessions by 2050 amounts to only just 2% of expected production – significantly less than the 12% reduction predicted in 1992 in the GSchG voting booklet. Political fears of massive electricity losses thus prove to be exaggerated.

People

Topics

  • Hydroelectric production and water protection
  • Residual water determinations (GSchG)
  • Energy transition and electricity supply
  • New concessions by 2050

Clarus Lead

A new residual water database systematically records for the first time 252 major hydroelectric power plants (≥3 MW) with their legal, hydrological, and technical attributes. The researchers thus analyze reliably how much residual water determinations reduce electricity production. The result defuses a central political debate: Instead of the feared massive losses from upcoming new concessions, simulations calculate only just 2% reduction by 2050. These figures allow the discussion between energy transition and water protection to be put on a factual basis.

Detailed Summary

Since the anchoring of residual water determinations in the Federal Constitution (1975, now Art. 76), there has been a fundamental conflict of objectives: How much water must flow in watercourses to protect ecology – and how much can be used for electricity production? Previous estimates of production losses were contradictory and based on incomplete data. Particularly problematic: It was nowhere publicly visible which residual water quantities were specifically stipulated.

The research team within the ETH program SPEED2ZERO has closed this gap. The database records 252 plants (160 run-of-river, 75 storage, 16 pumped storage, 1 circulation power plant) with a combined 31,540 GWh expected annual production – well over 84% of total hydroelectric production. For the first time, hydrological discharge time series from the catchment areas were considered to realistically model usable water, residual water, and overflow.

The simulation for 217 plants with currently defined residual water quantities results in an annual production reduction of 1,113 GWh (approximately 3% of domestic production). Crucial for the political debate: With strict application of residual water determinations for new concessions by 2050, an additional reduction of only 484–725 GWh results – that is less than 2% of expected production in 2050. Even with increased residual water releases through interest balancing (seasonally graduated), the total reduction by 2050 is only around 4.5%.

This stands in sharp contrast to the 12% reduction predicted in 1992. Swiss hydroelectric production has increased by over 4,000 GWh since 1992 through new construction and expansion as well as technical improvements – a compensation that is often overlooked in the current debate.

Key Statements

  • Previous estimates of production losses from residual water determinations were significantly too high and based on insufficient data.
  • New concessions by 2050 lead to an additional production reduction of just 2%, not the feared massive electricity losses.
  • The new database enables a factual debate between energy transition and water protection with reliable figures.
  • Since 1992, hydroelectric production has increased by over 4,000 GWh despite residual water determinations.
  • Approximately 70% of production reduction falls in the winter half-year – relevant for electricity supply security.

Critical Questions

  1. Data Quality & Completeness: The database covers 252 plants ≥3 MW, but only covers 84% of hydroelectric production. How representative are the results for smaller plants, and which data sources were used for the 35 plants without sufficient information?

  2. Methodological Robustness: How sensitive are the simulations to assumptions about discharge time series and hydrological scenarios? Were climate change effects on discharge volumes that could be relevant by 2050 taken into account?

  3. Conflicts of Interest in Interest Balancing: Article 33 GSchG allows increases in residual water quantities through interest balancing. How frequently is this applied in practice, and is the forecast based on realistic assumptions or optimistic scenarios?

  4. Causality & Comparability: The 1992 forecast of 12% reduction is compared with the 2026 forecast of 4.5% reduction. Are the assumptions and system boundaries comparable, or has the calculation methodology changed fundamentally, which would distort a direct comparison?

  5. Implementation Risks: The simulation assumes strict application of Art. 31 GSchG. How high is the risk that exemptions in practice lead to higher production losses than modeled?

  6. External Effects: The analysis focuses on electricity production. Were cost-benefit effects for aquatic ecology, fish stocks, or tourism quantified to provide a complete decision-making basis?

Sources

Primary Source: Press Release: Data-Based Residual Water Calculations – https://www.news.admin.ch/de/newnsb/tpvhsSquch8BoKfHc0tnJ (February 24, 2026)

Supplementary Sources:

  1. WSL, University of Bern, Eawag – Residual Water Database (SPEED2ZERO Project)
  2. Waters Protection Act (GSchG) – Art. 31, 33, 76 Federal Constitution

Verification Status: ✓ February 24, 2026

This text was created with the support of an AI model. Editorial Responsibility: clarus.news | Fact-Checking: February 24, 2026