Fifteen years after the Fukushima Daiichi nuclear disaster and almost three years after the controlled release of ALPS-treated water (Advanced Liquid Processing System) into the Pacific began on August 24, 2023, the debate surrounding the so-called "contaminated wastewater" remains highly emotional. The term "contaminated deuterium water" is catchy in the media but technically inaccurate: it is water containing tritium (³H, a radioactive isotope of hydrogen) that has been purified by ALPS of almost all other radionuclides. Tritium remains as the only relevant residual component because it is chemically identical to normal hydrogen and cannot be technically separated. By March 2026, over 18 batches had already been released, the most recent (18th batch) on March 6, 2026, with tritium concentrations well below the Japanese operational limit of 1,500 Bq/L – typically 200–300 Bq/L after dilution.
The International Atomic Energy Agency (IAEA) has been continuously monitoring the process since 2023. In its periodic reports (most recently September 2025 and ongoing 2026), it consistently confirms: the release complies with international safety standards, and the radiological impact on humans and the environment is negligible. Measurements of seawater and fish samples near the plant show tritium levels that are either below the detection limit (often < 3–10 Bq/L) or only minimally above the natural background of approx. 0.1–0.2 Bq/L. In most samples, the addition from the release is not even measurable due to the enormous dilution in the ocean (billions of cubic kilometers of water).
Technical Background and Release Process
The water originates from groundwater seeping into the damaged reactor buildings, from fire extinguishing water, and from rain. It initially contains various radionuclides. ALPS removes 62 nuclides (e.g., Cesium-137, Strontium-90, Cobalt-60) to levels below legal limits. Tritium (half-life 12.3 years) remains because it exists as tritiated water (HTO). Before release, it is diluted with seawater so that the concentration at the outlet is < 1,500 Bq/L – this is 1/40th of the WHO drinking water limit (10,000 Bq/L) and 1/7th of the Japanese drinking water standard. The annual release upper limit is 22 TBq of tritium, which is below the plant's previous operational limits.
Discharges are proceeding as planned until 2026: Each batch comprises approx. 7,800 tons, spread over weeks. The IAEA has conducted independent sampling (most recently February 2026 with experts from China, Korea, Russia, Switzerland), including seawater and fish. Results: No significant increase in tritium concentrations in the vicinity. Models predict a maximum increase of < 0.01 Bq/L in large parts of the Pacific over the next decades – well below natural fluctuations.
Medical and radiobiological assessment of tritium
Tritium emits only weak beta radiation (max. 18.6 keV, average energy ~5.7 keV). It barely penetrates the skin and, once ingested (drinking, eating), is rapidly distributed in the body – primarily in body water. The biological half-life is approx. 10 days, meaning the body excretes it quickly. The effective dose per becquerel ingested is extremely low: approx. 1.8 × 10?¹¹ Sv/Bq (ICRP value).
During discharge, exposure occurs almost exclusively through the marine food chain: fish eat plankton, humans eat fish. Organically bound tritium (OBT) in proteins, etc., is more persistent, but studies (e.g., on a Japanese flounder model in 2025) show that even under conservative assumptions, OBT in fish remains comparable to natural background levels. The estimated dose for the most highly exposed consumer (high fish consumption, proximity to the plant) is 0.001–0.03 µSv/year – this is 10,000–100,000 times below the public dose limit of 1,000 µSv/year and comparable to 1–2 hours of flight at high altitude (cosmic radiation).
Deterministic effects (threshold doses) are ruled out. Stochastic risks (cancer) are theoretically present according to the LNT model but statistically undetectable: The additional cancer risk is < 10?? per person – in a global population of billions, this would correspond to a few cases over decades, immeasurable amidst the natural cancer rate of ~40%.
Other radionuclides (e.g., C-14, Sr-90) are present in trace amounts but below limits after ALPS. Measurements in fish and seawater show no relevant increases. Critical studies (2024–2025) warn of bioaccumulation or synergistic effects with other stressors (warming, microplastics), but these remain hypothetical and contradict IAEA and monitoring data.
Environmental impacts on marine ecosystems
The Pacific massively dilutes the discharge. Models (2025) show that within 25 km of the outlet, tritium increases by a maximum of 0.1% of the natural background. Further out, the effect is zero. Fish and plankton samples show no damage. Comparison: Natural tritium sources (cosmic radiation, nuclear weapons tests in the 1950s–60s) caused higher exposures without measurable population damage.
Local fisheries are suffering economically from reputational damage (e.g., Chinese import ban), not from radiation. Prices for Fukushima fish have partially fallen, despite low readings.
Points of criticism and open questions
Critics (some studies 2024–2025) argue: Long-term effects of tritium on cell DNA or ecosystems are underestimated; ALPS does not completely remove all nuclides; global distribution via currents is possible. One paper (2025) estimates higher risks when considering bioaccumulation instead of pure diffusion. Others call for moratoria or alternatives (evaporation, geological storage).
The IAEA and the majority of experts consider these concerns exaggerated: The release is comparable to the routine operation of other nuclear power plants worldwide (e.g., France, Korea, China emit more tritium annually). No evidence of significant damage after almost three years.
Summary of current evidence (as of March 2026)
- No measurable increase in radiation levels in seawater or fish beyond natural fluctuations.
- Human dose: Negligible (< 0.03 µSv/year max.).
- No documented health effects or marine damage.
- IAEA monitoring: Continuously positive.
- Indirect effects (fear, economy) continue to outweigh direct radiological risks.
The release addresses a technical problem (lack of space for tanks) without a discernible radiological catastrophe. Nevertheless, skepticism remains justified – transparency and independent measurements are essential.
Verified link list (as of March 2026):
- https://www.iaea.org/topics/response/fukushima-daiichi-nuclear-accident/fukushima-daiichi-alps-treated-water-discharge
- https://www.iaea.org/newscenter/pressreleases/tritium-level-far-below-japans-operational-limit-in-18th-batch-of-alps-treated-water-iaea-confirms
- https://www.iaea.org/newscenter/pressreleases/iaea-report-confirms-japans-alps-treated-water-discharge-continues-to-meet-international-safety-standards
- https://world-nuclear.org/information-library/appendices/fukushima-radiation-exposure
- https://www.pref.fukushima.lg.jp/site/portal-english/en-moni-k.html
- https://www.unscear.org/unscear/en/areas-of-work/fukushima.html
