A river can look clear and still carry a burden invisible to the eye. It can also run tea-brown after rain and remain ecologically healthy. That is why Japan’s annual river announcement deserves to be read more carefully than the irresistible headline “the cleanest rivers.” On July 24, the Ministry of Land, Infrastructure, Transport and Tourism named 13 waterways in its top water-quality group for calendar year 2025. They stretch from the snow country of Hokkaido to the mountain basins of Kyushu. None was placed first. All 13 reached the same analytical floor.

The co-leaders are the Shiribeshi-Toshibetsu, Shokotsu, Saru and Satsunai in Hokkaido; the Arakawa in Fukushima; the Miyagawa in Mie; the Kitagawa in Fukui; the Ogamo and Tenjin in Tottori; the Takatsu in Shimane; the Niyodo in Kochi; the Kawabe in Kumamoto; and the Gokase in Miyazaki. Their shared distinction is precise: the average of annual-mean biochemical oxygen demand, or BOD, across each river’s eligible stations was 0.5 milligrams per liter, and the corresponding average of the stations’ BOD 75-percent values was also 0.5. That is the Environment Ministry’s reporting floor. Below it, the national table does not say which river is cleaner.

Japan’s other headline was broad but equally easy to misread. The ministry reported that 94 percent of applicable monitoring locations on nationally important Class-A rivers—837 of 887 sites—met their assigned environmental standard. It did not say that 94 percent of all Japanese rivers are pristine, nor even that 94 percent of the 160 rivers considered for the top group qualified. The numerator is monitoring sites; the benchmark varies with the designated use of the water body; and the main river indicator is BOD.

13 co-leadersEvery listed river reached the same 0.5 mg/L reporting floor; the ministry did not rank them 1–13
160 riversThe national pool considered for the “most favorable water quality” designation
837 / 887 sitesApplicable Class-A river monitoring locations that met their assigned environmental standard
94%A site-compliance rate, not the share of all Japanese rivers declared clean
1958The year the national government began systematic water-quality surveys on major rivers
20 yearsThe Kawabe River’s unbroken run in the top group, beginning with 2006 results

The 13 rivers, without a false podium

The official phrase, suishitsu ga mottomo ryōkō na kasen, is best rendered as “rivers with the most favorable water quality.” It is a threshold category, not a beauty contest. The ministry averages qualifying national-management stations, so the result describes monitored reaches under a defined method. It does not compare every spring, tributary, canal or prefecture-managed reach in the country.

RegionRiver and prefectureLandscape and continuity
HokkaidoShiribeshi-Toshibetsu (後志利別川)
Shokotsu (渚滑川)
Saru (沙流川)
Satsunai (札内川)		Four northern rivers shaped by snowmelt, forested headwaters and strong seasonal flows. The Shiribeshi-Toshibetsu has qualified seven years running; the Satsunai three and the Shokotsu two, while the Saru returned after two years outside the group.
TohokuArakawa (荒川), FukushimaA mountain-fed tributary of the Abukuma system—not Tokyo’s river of the same name. It has reached the top group for 16 consecutive years since 2010.
ChubuMiyagawa (宮川), MieFlows from the Kii Mountains toward Ise. The clean upper river and the historically more burdened Seta tributary show how one basin can contain sharply different water stories.
KinkiKitagawa (北川), FukuiA Wakasa-region river entering its fifth consecutive year in the national top group.
ChugokuOgamo (小鴨川), Tottori
Takatsu (高津川), Shimane
Tenjin (天神川), Tottori		Two linked Tottori waterways—the Ogamo is within the Tenjin system—and the Takatsu flowing from the Chugoku Mountains to the Sea of Japan.
ShikokuNiyodo (仁淀川), KochiThe river celebrated for “Niyodo Blue” returned for a second consecutive year, but the official distinction rests on samples, not color or tourism branding.
KyushuKawabe (川辺川), Kumamoto
Gokase (五ヶ瀬川), Miyazaki		Mountain rivers with the longest current streaks: 20 years for the Kawabe and 13 for the Gokase.
The 13 are tied at the laboratory’s reporting floor. The honest headline is not “Japan found a winner,” but “13 monitored river systems reached the cleanest band this method can distinguish.”

What BOD actually measures

BOD is an old but useful question posed to a bottle of river water: how much dissolved oxygen will microorganisms consume while breaking down biodegradable organic material? In Japan’s standard method, the sample is held for five days at 20 degrees Celsius. More sewage, food residue, manure or other readily decomposable material generally means more microbial activity, more oxygen consumed and a higher BOD. Low BOD therefore indicates a small organic-pollution load and better conditions for oxygen-dependent aquatic life.

The annual mean describes the typical level. The 75-percent value is a tougher compliance statistic. Monthly results are arranged from low to high; with 12 measurements, the ninth value is used. A river cannot earn the top label merely because a few superb samples pull down its average: both its station-average annual mean and its station-average 75-percent value must land at 0.5 mg/L.

But 0.5 is not a microscope of infinite resolution. It is the prescribed reporting lower limit. A reported 0.5 may mean the true concentration is at or below that level; it does not establish that every listed river contained exactly the same amount. It is like timing runners with a clock that displays no fractions below one second: several can share the recorded result even if nature has separated them by smaller margins.

What the national top group does—and does not—say
  • It does say: biodegradable organic pollution was extremely low at the eligible monitoring stations under the ministry’s BOD method.
  • It does not say: every reach was sampled, the water is safe to drink untreated, or swimmers face no microbial risk.
  • It does not directly measure: PFAS, pesticides, heavy metals, pharmaceuticals, microplastics, nitrogen and phosphorus loads, trash, sediment disturbance or ecological connectivity.
  • It is not a scenic ranking: color, transparency, forest cover, fish abundance and public affection are different observations.

What “Class-A” means—and what 94 percent means

In Japanese river law, “Class-A” is an administrative designation, not a water-quality grade. A Class-A river belongs to a system considered especially important for national land conservation or the national economy. A small tributary can be Class-A because it belongs to such a system. The label tells the reader who manages the river and why it matters institutionally; it does not mean “Grade A water.”

Environmental compliance is also not measured against one universal number. River reaches are assigned classes according to uses such as drinking-water supply, fisheries, swimming, agriculture, industry and basic environmental conservation. The national BOD limits range from 1 mg/L in AA waters to 10 mg/L in E waters, with accompanying standards for pH, suspended solids, dissolved oxygen and E. coli. A monitoring site “meets the standard” when it satisfies the target assigned to that water body.

Thus the 94-percent result is both substantial and bounded. It says 837 of 887 applicable sites met the local BOD standard for rivers, or the relevant COD standard for lake-like waters and reservoirs included in the accounting. It does not place all 837 sites in the same condition. One may meet a demanding AA target; another may meet a less stringent target set for a different use.

The rate was 97 percent in the 2024 report—863 of 889 sites—after 96 percent in 2023. The new 94-percent figure deserves attention, but one annual movement is not by itself evidence that three decades of improvement have reversed. River chemistry changes with rainfall, temperature, flow, construction, accidents, sample timing and the set of eligible stations. The useful questions are which 50 sites missed their targets, whether the same reaches recur, and what pollutants and flow conditions explain the misses.

A national recovery written in oxygen

The long view is remarkable. Japan began national-government water-quality surveys on major rivers in 1958, when rapid industrialization, urban migration and weak wastewater controls were transforming waterways into convenient conduits for waste. Annual nationwide compilation began in 1971. In that early baseline, 27 percent of Class-A river monitoring sites recorded annual-mean BOD above 5 mg/L—a level now associated with the more permissive end of the environmental classes, not a national norm.

One flashpoint came on Tokyo’s Edogawa in 1958, when discharge from a paper mill damaged fisheries and fishermen clashed with factory workers. The episode accelerated the old Water Quality Conservation Law and Factory Effluent Control Law. Those laws regulated only designated waters and were criticized as “chasing” pollution after damage appeared. Japan’s 1970 Pollution Diet replaced them with the Water Pollution Control Law, extending effluent regulation to all public waters and strengthening prevention. Environmental quality standards for water followed in 1971.

The legal architecture mattered, but pipes and treatment plants did the daily work. Sewerage expanded; factory effluent limits tightened; prefectures adopted stricter add-on standards; household wastewater entered policy; river managers aerated, dredged, diverted and restored troubled reaches. The national government began river-purification projects in 1969 and dam-environment projects in 1975. The Clear Stream Renaissance 21 program arrived in 1993. In 1997, a revised River Act formally placed “improvement and conservation of the river environment” alongside flood control and water use.

The Tama River became a visual emblem of the turnaround. In the 1960s and 1970s, synthetic detergent foam accumulated around weirs in a basin absorbing metropolitan growth. As sewer connections and effluent controls advanced, ayu sweetfish returned in large numbers by the late 1980s. The Ayase and Yamato rivers once recorded BOD 75-percent values above 30 mg/L. In 2025, the representative Yamato site registered 1.5 mg/L. Recovery did not restore an untouched past; it created a new urban river through regulation, infrastructure and persistent local work.

The latest report offers another recovery in miniature. At the Tokura station on the Ina River in the Yodo system, annual-mean BOD fell from 6.0 mg/L in 2016 to 2.5 in 2025—the largest ten-year improvement among nationally managed stations. That value is nowhere near the top group’s reporting floor, yet the 3.5 mg/L decline may represent the more consequential achievement: a burdened urban reach becoming materially healthier.

The stories inside the clean-water map

Four of the 13 rivers are in Hokkaido, where lower population density and expansive headwaters can reduce direct wastewater pressure. But geography is not immunity. Dairy farming, tourism, dams, erosion, road runoff and wastewater systems still determine what reaches the channel. Hokkaido’s own 2025 accounting found 85 of 93 applicable environmental-standard sites compliant. Seventeen of its 20 assessed rivers had river-wide mean BOD below 1.0 mg/L—strong performance, but not universal perfection.

Fukushima’s Arakawa tells a different story. It descends steep volcanic terrain west of Fukushima City before joining the Abukuma. Its 16-year streak depends not only on mountain water but on people protecting the reach where mountains meet a city. Kumamoto’s Kawabe, a Kuma River tributary, has now spent 20 years in the cleanest band, through flood, dam controversy and intense debate over what a living river should provide. Continuity is impressive precisely because rivers are not laboratory plumbing.

The Miyagawa basin warns against letting a headline flatten geography. Its upper river is nationally celebrated, while the Seta River flowing through Ise historically carried a heavier urban organic load. The previous national report identified Seta Bridge as the country’s largest ten-year BOD improvement. The basin therefore contains both a top-group main river and a tributary remembered for remediation. “Clean river” is never a permanent adjective for every branch.

The Niyodo shows another distinction. Its blue-green clarity has become a regional identity and visitor economy. Yet optical beauty and BOD answer different questions. Glacial flour can make pristine water opaque; dissolved contamination can pass through crystal water unseen. The strongest river stewardship treats the tourism image as an invitation to examine more evidence, not as a substitute for it.

Why 13 is fewer than last year’s 20

The number of top-group rivers fell from 20 in the 2024 survey to 13 in 2025. Seven fewer names makes an arresting trend line, but the category has a hard edge at the reporting floor. A small change from 0.5 to 0.6 mg/L in either of the two averages removes a river from the list even though its water remains exceptionally low in organic pollution. Conversely, returning to 0.5 can restore the designation.

Weather can matter. Heavy rain washes soil, organic matter and urban residue into channels; low flow can reduce dilution; warmer water accelerates biological processes and holds less oxygen. Sampling schedules and station coverage matter too. The national report rightly highlights sustained streaks because a decade or two at the floor says more about resilience than a single appearance. It does not, however, publish a causal diagnosis for every river that entered or left the annual group. Claims that the decline proves nationwide deterioration would outrun the evidence.

Beyond chemistry: insects, senses and citizens

A healthy river is more than a low BOD result. Since fiscal 1984, river managers and the Environment Ministry have invited students and residents to examine aquatic insects whose tolerance to pollution integrates conditions over time. In 2025, public-participation investigations covered 84 river systems and 268 sites in the ministry-reported program. Stoneflies, caddisflies, mayflies and other indicator organisms turn a technical standard into a living question: what can survive here?

People also record rubbish, smell, transparency and the feel of the riverbed. These observations are not second-rate substitutes for a laboratory. They capture dimensions the BOD bottle misses and place witnesses beside the water often enough to notice a fish kill, oil sheen, illegal dumping or new source of turbidity. In the Chugoku region alone, 50 groups with 1,767 participants examined 49 sites; 96 percent of the investigated sites fell into the two cleaner biological categories.

The categories can disagree. A reach may have low organic pollution but poor habitat because concrete banks, barriers and depleted gravel leave little room for life. Another may support rich habitat while a health pollutant requires targeted chemical testing. Good governance keeps those signals separate long enough to understand them, then brings them together in a basin-wide decision.

The next generation of river risk

Japan’s postwar water-quality system was built to confront visible industrial discharge and oxygen-robbing organic waste. Those problems have not vanished, especially in urban tributaries, livestock areas and enclosed lakes, but the risk portfolio is expanding. PFOS and PFOA became enforceable drinking-water quality items in Japan on April 1, 2026. Monitoring for persistent chemicals, pharmaceuticals and microscopic plastics demands different instruments, time scales and regulatory responses from BOD.

Climate volatility complicates the picture. Intense rainfall can overwhelm sewers, mobilize sediment and move pollutants rapidly downstream. Drought concentrates contaminants and warms shallow water. Forest management, agricultural practice, groundwater, dams and the quantity of flow can become as important as the concentration measured in one bottle. The ministry’s move toward integrated basin management recognizes the basic truth: water quality cannot be separated from water quantity, land use, flood control and habitat.

There is also a fiscal challenge hidden inside the success. Mature sewer networks and treatment plants require renewal. Rural depopulation can leave fewer ratepayers supporting aging systems, while popular clear-water destinations must manage visitors, toilets, litter and river access. The cleanest rivers are not free gifts from empty landscapes. They are assets with maintenance costs and with upstream communities doing work that downstream users may barely see.

How to read the river report honestly

First, celebrate the 13. Reaching the reporting floor at multiple stations, through seasonal variation, is evidence of extraordinarily low biodegradable organic pollution. The long runs on the Kawabe, Arakawa and Gokase show that this condition can be defended over decades rather than discovered for a summer.

Second, resist turning the list into a license. Do not drink untreated river water because of a BOD ranking. Do not infer that a beautiful river is free of pathogens or persistent chemicals. Do not call an unlisted river dirty: it may have measured 0.6 mg/L, may fall outside the national target pool, or may be managed by a prefecture rather than the national government.

Third, watch the sites that missed their standards and the rivers that are improving from a difficult baseline. National environmental progress is not only the preservation of famous blue water. It is also the unglamorous reduction of sewage, grease and organic loads in the working tributary behind a station, warehouse or apartment block.

Japan’s 2025 result is best understood as a layered achievement. Thirteen rivers reached the lowest BOD band the national table can resolve. Most monitored Class-A locations met the standards assigned to them. That outcome rests on a history that includes poisoned fisheries, public conflict, stronger law, vast sewer investment, citizen sampling and half a century of patient measurement. The river at the end of that history is neither pure nor finished. It is watched—and because it is watched, it can still be defended.

Sources and methodology

The national results cover calendar year 2025 and were released on July 24, 2026. “Most favorable” is the ministry’s BOD-based category, not a drinking-water, scenic or comprehensive ecological ranking.