Provisional data. Figures on this page are preliminary and subject to revision as monitoring continues. The water-temperature data is the Karuk Tribe’s measured record at Iron Gate, used with the Tribe’s permission.

Klamath Irrigation District · Setting the Record Straight

The Facts on Spring Flows and Young Salmon

Updated · June 21, 2026

Update — June 2026: what really happened to the young Chinook

The short version. In early June, large numbers of young Chinook salmon turned up dead in the upper Klamath River. They were almost entirely hatchery fish — part of the 675,000 young Chinook California released on May 14 from the Fall Creek Hatchery, about 7 miles above the old Iron Gate Dam site and roughly 200 miles from the ocean.

They went into the upper river in mid-May, and — according to the Oregon State University scientist examining them — instead of heading for the ocean they lingered for weeks right where a natural fish parasite (C. shasta) concentrates. As the water warmed and parasite levels climbed, sitting in that reach is what made them sick. Fish that move down to cooler water and reach the sea can shed the infection.

The wild run had mostly passed through weeks earlier, while the water was still cold, and largely escaped the parasite. And the threatened fish the river’s flow rules are written to protect — coho salmon — live in cool tributaries, not the warm summer mainstem where this happened.

The bottom line

  • These were hatchery Chinook — not the wild run, and not the threatened coho the flow rules exist to protect.
  • Their deaths trace to where and when they were released: far upstream, into warming water, in mid-May. They lingered there for weeks — where a natural parasite builds up — instead of heading to sea.
  • 2026 was not a low-flow year. The river ran well above the drought of record through the spring.
  • The extra water being demanded is warm lake water. It feeds the parasite; it does not cool the river.

Three different fish — and the one that died isn’t the one the rules protect.

Coho salmon

Oncorhynchus kisutch

ESA-listed · threatened

The species the spring flow rules exist to protect. Coho live in cool tributaries and stayed essentially absent from the warm summer mainstem all season.

Not part of this die-off.

Wild Chinook

Oncorhynchus tshawytscha · natural-origin

Not ESA-listed

About 580,000 counted this season; an estimated 80% had left the river by early April, while the water was still cold.

The wild run largely escaped.

Hatchery Chinook

Oncorhynchus tshawytscha · Fall Creek Hatchery

Not ESA-listed

675,000 released May 14, far upstream into already-warm water; they lingered for weeks where the parasite concentrates.

~2,170 documented dead in the traps — the fish in the news (a floor).

Three facts worth keeping in view

  • The fish the rules protect weren’t there. The spring flow rules exist under the Endangered Species Act to protect threatened coho salmon. Coho stayed essentially absent from the mainstem traps all season — they live in cool tributaries, not the warm summer mainstem where this die-off happened.
  • This tracks hatchery decisions, not flow. The 2026 mortality appears tied directly to where and when hatchery fish were put into the river: 675,000 smolts (young Chinook ready to migrate to sea) released May 14, far upstream, into already-warm water.
  • The whales are downstream of all of it. Of the roughly 75 endangered Southern Resident killer whales, it is L pod — about 34 — whose ocean range carries it past the Klamath in winter; J and K pods take little to none of the basin’s Chinook. Three to four years from now, that pod will have numerous fish to draw on, including the ones that survive from this year’s full outmigration (the season’s young fish heading to sea) — wild and hatchery alike — which points to the survival of every juvenile, not the volume of water sent past the headgate.

A fuller account of the die-off

On the claim that low flows caused this

Some recent coverage says low river flows caused the die-off, and that releasing more water would have prevented it.

We don’t dispute this is a warm-water-and-parasite event; in a dry, low-snowpack year, flows and warm water tend to arrive together naturally. What matters is the human decision to release additional warm water onto existing, known conditions.

The operative variable is temperature — and 2026 was not a low-flow year. A low-snowpack winter warms the river regardless of how much water is released, and it is the warmth and parasite dose that harm the fish, not the volume by itself. By the measured record (the chart further down), 2026’s Keno releases ran well above the drought of record — several times that benchmark’s flow through the spring — so the river was not running low when these fish died.

The water being demanded is warm. The additional flow being demanded — including under the 2024 Biological Assessment and Opinion — would come from Upper Klamath Lake, and by late spring that lake water is warm. Federal monitoring identifies warm water as the dominant driver of this parasite. Sending more lake water downstream adds warm water — it does not cool the river or remove the parasite. The demanded remedy delivers more of the very condition that feeds the disease.

The timing and the fish don’t fit. These were late-released hatchery fish that stayed in the warm upper river; the wild run had already gone downstream in cold water. A larger June release of warm stored water would arrive after the wild fish have passed, would not move hatchery fish that aren’t migrating, and cannot cool a river warmed by the summer sun.

What would actually help is a shorter, cooler trip. Releasing hatchery fish lower in the river and earlier — closer to the cool ocean, before the water heats up — shortens the warm-water gauntlet. That is a hatchery decision, not a question of how much water is sent past the farms.

~2,170young Chinook documented dead or dying in the Iron Gate & I-5 traps, June 2–5 — a floor, since traps catch only part of the run
~0.3%of the 675,000 hatchery smolts released May 14, documented dead in those traps — though a majority of the fish actually caught were dead (~60%, up to ~80% at some sites)
~0%of the dead were wild fish — the ~580,000 wild run had already moved downstream in cold water

Here is the same story in more detail, with the monitoring data behind it. (The two timelines further down now run through June 21.)

What the traps found. In the week of June 2–5, the U.S. Fish and Wildlife Service reported that about 60% of the juvenile Chinook caught at the Iron Gate and I-5 traps were dead or dying (60% of unmarked fish, 68% of ad-clipped), with more dead fish along the banks nearby; some upper-river sites reported up to ~80%. Of roughly 3,500 Chinook caught at those two traps, an estimated ~2,170 were dead or dying. By their fin-clips, nearly all were hatchery fish — part of the 675,000 smolts the Fall Creek Hatchery released on May 14. Most were found between Fall Creek and the Shasta River. That ~2,170 is a documented floor, not a river-wide total: the traps sample only a fraction of the fish moving past.

These were not the wild fish in the timelines below. That run had largely passed weeks earlier, in cold water. These were Fall Creek Hatchery fish, released in mid-May into a mainstem that was already warm.

An infection rate is not a death rate.

A fish that tests positive for C. shasta is infected — not necessarily dead. The parasite can be detected within about a week of exposure, but the mature spores that actually kill a Chinook take two to three weeks to develop. A high detection number measures exposure, not a body count.

And exposed fish can recover. C. shasta is a freshwater parasite; it cannot complete its life cycle in the sea. Chinook that reach cooler water downstream and make it to the ocean can shed the infection and recover — as the fish pathologist monitoring this event reiterated at the June 11 operations meeting. What decides the outcome is not infection alone; it is temperature, and how far the fish must travel through warm water.

That fits what the biologists examining this event described: the released hatchery fish did not head downstream promptly. They lingered for weeks near the release reach — exactly where the parasite is most concentrated — and took a heavy dose. Had they moved to cooler water sooner, they would have met far less parasite and reached the ocean, where the infection cannot survive (Oregon State University).

What CDFW reported

  • CDFW timed the May 14 release for a window when C. shasta in the river had eased, and ahead of forecast storms expected to improve water quality and speed the smolts to the ocean. C. shasta then climbed, and some of the released fish were later found dead in the traps near the former Iron Gate Dam.
  • Lab work confirmed two parasites in the dead fish: C. shasta and Parvicapsula minibicornis.
  • Most wild juveniles had already migrated out ahead of the rising parasite levels, and CDFW says it remains confident that a share of the released hatchery fish escaped the parasites and reached the sea — where C. shasta does not affect the fish or ocean fishing.
  • Some pathogen loss is a normal part of the Chinook life cycle: a pair of Chinook produce roughly 4,000 offspring, and even under ideal conditions more than 99% die to disease or predation before reaching adulthood.
  • CDFW also notes that early-spring 2026 C. shasta ran lower than in pre–dam-removal drought years, and that C. shasta mortalities in spring 2024 and 2025 were significantly lower — which it calls encouraging signs.

Two parts of that account are worth setting against the record. First, the expectation that forecast storms would speed the smolts to the ocean is not what the outmigration science finds: migration rate is largely unaffected by flow, and a juvenile’s survival tracks the distance it must travel, not how fast the water moves it (Muir et al. 2001; Smith et al. 2002; Williams et al. 2005, reviewed in Lestelle 2007 — quoted below). Neither the forecast storms nor a larger June release would have carried these fish to sea fast enough to change the outcome. Second, while 2026’s spore densities ran below the 2021 drought-year peak, the parasite did not ease in the corridor these fish had to cross — Beaver Creek read 39 spores/L on May 11, three days before the release. By June 1, Oregon State University’s water-column monitoring had four of six mainstem index sites above 10 spores per liter — the density that program ties to a 40% mortality threshold in juvenile Chinook (KSH 22, KBC 17, KMN 14, KSV 13 spores/L; only the two lowest sites fell below the line). That was the parasite level the released fish were moving through in the days just before the June 2–5 die-off — not an eased river.

That points to the part of this that is actually within human control: distance. Mortality concentrates where juveniles spend the most time in warm, infectious water — and Fall Creek Hatchery sits about 200 river miles from the Pacific — roughly 7 miles above the former Iron Gate Dam site — so its fish must run that entire corridor. This is not our characterization. In 2021, the California Department of Fish and Wildlife trucked about 1.1 million hatchery juveniles toward the coast rather than releasing them upstream. In the agency’s own words, the science indicated the fish would not survive the journey — up to ~75% projected mortality over the ~192 river miles.

This concern is not new. CDFW’s own review of Iron Gate Hatchery records a long-standing recommendation to release fewer smolts and more fall yearlings — specifically to avoid the hatchery–wild interactions and mortality that occur when juveniles are in the mainstem during the low-flow, poor-water-quality conditions of late spring and summer (Joint Hatchery Review Committee 2001, in California Department of Fish and Game, Chesney 2007). The 2026 fish met that window less by their May 14 release date than by lingering in the warm upper reach instead of moving downstream.

The research is blunt on this point:

“Migration rate is unaffected by flow; survival appears to be largely a function of migration distance and not travel rate.”Muir et al. 2001; Smith et al. 2002; Williams et al. 2005 (reviewed in Lestelle 2007)

In other words, more flow does not move juveniles to the sea fast enough to change the outcome — the distance itself is the risk, and a longer trip means more predator encounters along the way. That makes one post-dam-removal decision hard to square with the science: the Klamath River Renewal Corporation moved hatchery production farther upstream, lengthening the very corridor that drives mortality. And the new Fall Creek Hatchery is built overwhelmingly for Chinook — its coho production goal is just 75,000 fish, about 2.3% of the planned 3,325,000-fish annual release (Federal Energy Regulatory Commission, Draft EIS FERC/EIS-0313D, Feb. 2022). Releasing millions of hatchery Chinook into the upper river raises a fair question about added competition for the very coho the spring flows are meant to protect.

The June 12 disease signs map onto that distance directly — high through the warm upper and middle river, and gone by the site nearest the ocean:

Sick upstream, clean near the sea

Juvenile Chinook with visible disease signs (pale-or-worse gills), by river mile · week of June 9–12, 2026

200 mi 150 100 50 0 mi · sea 0 10 20 30 40 50 % of fish with disease signs Fall Creek Hatchery — release point 675,000 smolts, May 14 · ~7 mi above the old Iron Gate Dam 44.9% 48.9% 49.1% 0% 0 of 70 fish Iron Gate I-5 Kinsman Weitchpec 193 mi 182 mi 148 mi 40 mi river miles from the Pacific upper river (warm) direction of migration → toward the ocean
Source — USFWS Arcata Fish & Wildlife Office Outmigrant Update, June 12, 2026 (Table 4), plotted against river miles from the Pacific. Disease-sign sampling is conducted at the four mainstem trap sites shown; the fish originate at the Fall Creek Hatchery, about 7 miles above the former Iron Gate Dam site (~200 river miles from the sea), which released 675,000 Chinook smolts on May 14 (CDFW). Whether the clean lower-river fish shed the infection or the sickest did not survive that far, the burden falls on the long upper-river run. Visible signs are an index, not a mortality count. Provisional.

And the listed species? Coho remained essentially absent from the mainstem traps through this entire window — as they have all season. The fish in the warm-water tail were Chinook, which are not listed. The threatened coho that the spring flows are legally meant to protect are a tributary fish, not present in the warm summer mainstem where this disease plays out.

None of this means the loss was nothing. It means the loss fell on late-released hatchery fish, in a warm reach, far from the sea — and that the lever most directly tied to it is where and when those fish are put in the river, not how much stored Project water is sent past the headgate in June.

None of this rests on KID’s reading of the data. The monitoring behind it runs through the Klamath Fish Health Assessment Team (KFHAT) — the basin’s joint fish-health body, whose members include the Yurok and Karuk Tribes, NOAA Fisheries, CDFW, USFWS, and the Bureau of Reclamation. KFHAT’s in-season 2026 group reports are the shared, multi-party record of this event, and its public readiness map places the flag for elevated juvenile-Chinook mortality in the upper reach (roughly Fall Creek to the Scott River) while rating most of the Klamath–Trinity basin favorable — with the Scott, Shasta, and Salmon tributaries at a watch level for warm temperatures and low flows. That is the same localized pattern the river-mile chart above shows (Klamath Fish Health Assessment Team, 2026 in-season group reports and readiness map; Klamath Basin Monitoring Program).

An open question for a post-dam river

If a “flushing flow” is warm water from a shallow lake rather than cold water from snowmelt, does the thermal acceleration of the parasite cancel out the physical scouring of its host?

With the dams gone, the river’s channel dynamics are new. The 2026 die-off suggests an old rule of thumb — that “more water” is automatically better for fish — deserves a careful, data-driven look rather than assumption. The alternate hypothesis rests on basic thermodynamics and local hydrology:

  1. Upper Klamath Lake is a heat trap. It is a very large, very shallow basin. In a low-snow year with strong spring sun, it warms quickly. Releasing that water does not mimic a cold snowmelt pulse — it delivers warm, nutrient-rich water into the mainstem.
  2. Cold springs are the river’s natural air-conditioning. The reach below Keno and into the canyon is fed by cold-water spring complexes that hold a chilled baseline. Documented Klamath thermal refuges have been shown to reduce juveniles’ exposure to C. shasta — cool-water patches at tributary confluences carry lower parasite concentrations and ease disease progression (Chiaramonte et al. 2016, Trans. Am. Fish. Soc. 145:810–820).
  3. The two can work against each other. Pushing warm lake water downriver to hit a flow target during a drought could overwhelm that localized spring cooling. The added warmth can speed the parasite’s development, while higher, faster flow can mix out the cold-water pockets juveniles use as refuge — the opposite of the intended effect.

Why temperature is the hinge: in Klamath salmonids, C. shasta-induced mortality rises steeply as the water warms. So if a given flow target can be met either with warm lake water (pushing the mainstem warmer) or by leaning on the cooler, spring-fed reaches, the temperature difference may matter more to survival than the added volume. That is a testable question, not a settled answer — and it is the kind of question a post-dam Klamath now demands.

On the temperature–disease relationship: Ray, R.A., Holt, R.A. & Bartholomew, J.L. (2012), Relationship between temperature and Ceratomyxa shasta–induced mortality in Klamath River salmonids, Journal of Parasitology 98(3):520–526.

Each spring, we hear that large river releases are needed to protect juvenile fish from disease. This year's monitoring tells a more specific story. The chart below puts three things on one timeline: when the wild young Chinook leave the river, how warm the water gets, and how the C. shasta infection rate climbs.

What this shows: the wild young Chinook had largely left the river — the tall early peak — before the water warmed and the C. shasta infection rate climbed in late April and May. The fish caught in that warm-water window were mostly the late-released hatchery fish, not the wild run. The gold line marks when those hatchery fish were released — May 14, 675,000 Chinook from the Fall Creek Hatchery. By the June 5 trapping update, the natural-origin (wild) catch had fallen to its season low, and the fish still showing up at the Iron Gate and I-5 traps were mostly hatchery releases. The red diamonds track only distended bellies; a second warm-water parasite hit harder still — at the Iron Gate trap, most young fish sampled in mid-May had pale or diseased gills.

Temperature — the Karuk Tribe’s measured record at Iron Gate (Klamath River below Iron Gate), daily mean/min/max through June 21, used with the Tribe’s permission (provisional). Coverage — temperature runs through June 21; the Chinook catch and the distended-belly visual sign run through June 4 (latest USFWS trapping update, the week of June 2–5). The qPCR C. shasta prevalence markers run through May 7; the newest qPCR infection rate is pending the California–Nevada Fish Health Center’s next update. Chinook — natural-origin (wild) weekly catch, all reporting traps, digitized and scaled to published season totals. C. shasta — infection rate in weekly corridor samples; “est. infected” couples that rate to the catch. Markers show prevalence (the share of fish infected), not mortality. Pale gills — a separate sign of anemia linked to Parvicapsula minibicornis, a second temperature-driven parasite (not plotted). At the Iron Gate trap the share of sampled fish with pale-or-worse gills ran 70.8% (May 5–8) and 50.5% (May 12–15), easing to 8–23% through early June (USFWS Table 4).

What the data shows

Ceratonova shasta (C. shasta) is real, and we take it seriously. It is also seasonal — it tracks water temperature. This year the first detection came on April 2, and the infection rate in the weekly samples then climbed steeply, reaching roughly 90–100% by late April as the river warmed.

By then, most of the wild run was already gone. Across all the reporting juvenile traps — Shasta, Scott, Bogus, Fall, Shovel, Jenny, and the ODFW Spencer and Klamath stations — about 580,000 natural-origin young Chinook were counted this season, and an estimated 80% had emigrated before that first April 2 detection, while the water was still cold.

In plain terms: the bulk of the wild outmigration moved through cold water, ahead of the disease-and-temperature spike. We do not pretend the cost was zero — coupling the catch to the measured infection rate puts an estimated 13% of the run (about 75,000 fish) as infected, mostly in the warm-water tail. But the largest spring flow demands arrive after most of the fish they are meant to protect have already passed the upper river.

The water-column monitoring, week by week

The parasite dose itself — the density of C. shasta spores in the river water — is tracked separately by Oregon State University at six mainstem index sites, and it is the clearest measure of what an outmigrating fish actually swims through. The full 2026 record tells the story the snapshots cannot: the dose stayed concentrated in the upper river — the reach nearest where the hatchery fish were released — and it was already above the mortality threshold there when those fish went in on May 14, not eased.

What this shows: water-column C. shasta spore density at all six index sites across the 2026 season — the parasite dose an outmigrating fish actually swims through, week by week. Read it against the hatchery timeline: the fish were released on May 14, by which point the upper corridor had climbed back above the 10 spores/L line the program ties to a ~40% mortality threshold in juvenile Chinook (Beaver Creek, KBC, measured 39 sp/L on May 11, three days before release). The upper sites — the reach the released fish had to cross — stayed at or above that line through the outmigration and the June 2–5 die-off, with the Shasta-confluence site (KSH) reaching the year’s highs of 56 (May 26) and 66 sp/L (June 8), even as the lower river (KOR, KTC) stayed low. The dose was heaviest exactly where and when the released fish were present.

Source — Oregon State University, Department of Microbiology (Bartholomew Laboratory), Klamath C. shasta water-column spore monitoring, weekly 2026 updates (six mainstem index sites, late March through June 15), read from the program’s posted per-site spore densities. Most weeks are 24-hour composite (ISCO) samples; a few are manual grab samples, which are more variable (e.g., the low KBC reading on May 26). What this is — spore density is the parasite dose in the water, which is different from the share of fish infected shown in the timeline above. The 10 spores/L value is the program’s own reference line. Genotype — the program also types the parasite. The coho-associated form (Type II, which it calls the “coho” genotype) stayed at trace levels (<1 spore/L) or undetected at every site, every week, all season — against a total load that reached 66 spores/L. The heavy dose in the corridor was the non-coho, Chinook-affecting type throughout.

Regulated for one fish — while the water follows another

Here is the part worth sitting with. The flows that limit water to Klamath Project farms are required under the federal Endangered Species Act to protect coho salmon, which are listed as threatened. Coho are the listed species — they are the legal reason the river is given priority over the headgate in a dry year like 2026, when stored Upper Klamath Lake water is released to hold river flows above what the season would naturally provide.

But the fish in this chart are Chinook — and Klamath Chinook are not listed under the federal Endangered Species Act, fall run or spring run. They are the fish that fill the juvenile trap counts, and they are the fish at the center of the disease monitoring used to argue for higher spring flows.

There is one more link in that chain. Klamath Chinook still enter the federal picture — not on their own, but as food. Reclamation's own 2024 Biological Assessment defines an action area that reaches out into the Pacific Ocean, where, in its words, the Southern Resident Killer Whale — an endangered population of orcas — “feed on concentrations of adult Chinook Salmon.” Chinook are the orca's primary prey. So the unlisted Klamath Chinook are folded into the federal consultation indirectly: as ocean prey for a listed animal that lives hundreds of miles away. Whether that distant link should drive how a dry-year river is rationed at the headgate is a fair thing to put on the table.

So the record raises a fair question: Klamath farmers are being curtailed for a listed species (coho), while the disease case used to argue for higher spring flows is built largely around an unlisted one (Chinook). And when the bulk of those Chinook have already passed in cold water — as the chart shows — it is fair to ask how the late-spring flows line up with the fish they are meant to protect, and which fish those are.

There is also a gap the record never tested. Two separate questions get blurred together: what the Endangered Species Act actually requires to avoid jeopardizing coho, and how much water is demanded above that legal floor. A biological opinion — the federal review that decides whether a water plan complies with the Act — evaluated operations that were already curtailed, and found them ‘no jeopardy.’ Deliveries nearer the Project’s own water rights were never put on the table to test. So the volume demanded in a dry year has never had a clean hearing against what the law strictly requires.

A closer look: where the 2026 water came from

This second chart uses Reclamation's own daily accounting. The blue line is the natural water arriving at Upper Klamath Lake (net inflow — the natural water arriving; the A Canal is already accounted for in the lake’s water budget, not subtracted from inflow corrected 6/12); the orange line is the flow released downriver at Keno to meet the federal target; the gold line is the lake's surface elevation. Through the winter and the cold-water outmigration — when the young Chinook were actually leaving — inflow ran well ahead of the release, and the lake filled more than three feet, peaking on April 5. Only after that, as natural inflow fell away, did the release begin to outrun it: on 57 of the 77 days after the April 5 peak, the prescribed Keno flow exceeded what nature delivered to the lake that day, so the river was held up by drawing stored water back out. In plain terms, the stored water that sustains the prescribed flow is spent mostly in May and June — after the fish have gone.

And the “low flows” framing does not match the record. The two dashed lines put 2026 in historical context. The two gray lines are prior years for comparison: the dashed line is the dam-removal water year (2024), and the fainter dotted line is water year 1992 — the drought of record, from the USGS Keno gage. The 1992 line sits below 2026 in every month; 2024 runs lower through the spring and June, though it pulsed higher than 2026 for a few weeks in late winter. In 1992 the river at Keno never exceeded about 670 cfs all year, and ran at roughly 150–200 cfs through the spring and into June — the very months the young fish move. The 2026 releases (orange) run at or above the dam-removal year and several times higher than the 1992 drought across that same outmigration window. Whatever else is debated about how the water is managed, the measured record does not show 2026 as a low-flow year.

Net inflow & Keno release — Reclamation daily UKL accountability, WY2026 (net inflow = storage change plus Link River and A Canal releases, back-calculated from the lake’s water budget; provisional — corrected 6/12). UKL elevation — Reclamation daily surface elevation (ft); series through June 21. Keno release — 2024 — the dam-removal water year (Oct–Jun), overlaid on the same calendar for comparison (Reclamation ReclamationTracker, col I). Keno flow — 1992 — the drought of record, USGS Keno gage 11509500 (approved daily discharge), Oct–Jun on the same calendar; it never exceeded ~670 cfs and fell to ~150 cfs by June. Deficit days — days the Keno release exceeded same-day net inflow; 57 of those fell after the April 5 storage peak.

A few notes, because facts cut both ways: this chart tracks Chinook, not coho, and does not measure coho directly; the Chinook counts are natural-origin (wild) fish, digitized from agency briefing charts and scaled to each trap's published season total; the infection figures are the rate found in small weekly corridor samples (about 30 fish each, wide margins of error), and the “estimated infected” line couples that rate to the catch — it is an estimate, not a head count; temperatures are the Karuk Tribe’s measured record at Iron Gate, used with the Tribe’s permission. All figures are provisional and subject to revision.

We support healthy fish runs. We also believe water decisions affecting hundreds of farm families should rest on measured timing and real numbers. The cold-water window is when these fish actually move — and that is where management should follow the science.

For the broader policy case — that Basin recovery turns on temperature, water quality, and disease as much as water volume, and on integrated approaches that serve fish, farms, refuges, and wildlife together — the Klamath Water Users Association makes that argument in its companion release, Beyond Water Volume in the Klamath Basin (June 2026).

Sources

  • Ray, R.A., Holt, R.A. & Bartholomew, J.L. (2012). Relationship between temperature and Ceratomyxa shasta–induced mortality in Klamath River salmonids. Journal of Parasitology 98(3):520–526. doi.org/10.1645/JP-GE-2737.1
  • Chiaramonte, L.V., Ray, R.A., Corum, R.A., Soto, T., Hallett, S.L. & Bartholomew, J.L. (2016). Klamath River thermal refuge provides juvenile salmon reduced exposure to the parasite Ceratonova shasta. Transactions of the American Fisheries Society 145(4):810–820. doi.org/10.1080/00028487.2016.1159612
  • Muir, W.D., Smith, S.G., Williams, J.G. & Sandford, B.P. (2001). Survival of juvenile salmonids passing through bypass systems, turbines, and spillways with and without flow deflectors at Snake River dams. North American Journal of Fisheries Management 21:135–146. nwcouncil.org
  • Smith, S.G., Muir, W.D., Williams, J.G. & Skalski, J.R. (2002). Factors associated with travel time and survival of migrant yearling Chinook salmon and steelhead in the Lower Snake River. North American Journal of Fisheries Management 22(2):385–405. (Full text by subscription through the journal.)
  • Williams, J.G., Smith, S.G., Zabel, R.W., Muir, W.D., Scheuerell, M.D., Sandford, B.P., Marsh, D.M., McNatt, R.A. & Achord, S. (2005). Effects of the Federal Columbia River Power System on salmonid populations. NOAA Technical Memorandum NMFS-NWFSC-63. nceas.ucsb.edu
  • Lestelle, L.C. (2007). Coho Salmon (Oncorhynchus kisutch) Life History Patterns in the Pacific Northwest and California. Final report prepared for the U.S. Bureau of Reclamation, Klamath Area Office. coastcoho.org
  • California Department of Fish and Game. Iron Gate Hatchery review (Chesney 2007), including the Joint Hatchery Review Committee (2001) release-timing recommendation. nrm.dfg.ca.gov
  • California Department of Fish and Wildlife. Understanding juvenile salmon mortality in the Klamath River (statement, June 20, 2026; via Maven’s Notebook). mavensnotebook.com
  • Klamath Fish Health Assessment Team (KFHAT) — 2026 in-season group reports and readiness map. Klamath Basin Monitoring Program. kbmp.net/collaboration/kfhat
  • Oregon State University, Department of Microbiology (Bartholomew Laboratory). Klamath River Ceratonova shasta water-column spore monitoring — weekly 2026 updates, six mainstem index sites, late March through June 15 (the 10 spores/L reference line corresponds to a 40% mortality threshold in juvenile Chinook). microbiology.oregonstate.edu
  • U.S. Geological Survey streamgage 11509500, Klamath River at Keno, OR (water year 1992 daily discharge). waterdata.usgs.gov
  • Federal Energy Regulatory Commission (2022). Final Environmental Impact Statement, Lower Klamath Project (FERC Project Nos. 14803-001 and 2082-063; eLibrary accession 20220826-3006, August 26, 2022). elibrary.ferc.gov
  • National Marine Fisheries Service (2024). Endangered Species Act Section 7(a)(2) Biological Opinion for Klamath Project Operations, 2024–2029 (NMFS No. WCRO-2024-01599; October 28, 2024). usbr.gov
  • U.S. Fish and Wildlife Service, Arcata Fish and Wildlife Office. Klamath River Juvenile Salmonid Outmigrant Monitoring Update — June 5, 2026 (weekly trap catch and the early-June fish-health figures). fws.gov
  • Klamath Water Users Association. Beyond Water Volume in the Klamath Basin: Realizing Solutions for Fish, Farms, and Wildlife (informational release, June 23, 2026) — the Klamath Project water-users’ companion policy statement on temperature, disease, and integrated Basin solutions. kwua.org
  • Water temperature data: Karuk Tribe Water Quality Program final record at Iron Gate, used with the Tribe’s permission (not a public dataset).
  • Flow and lake-elevation data: U.S. Bureau of Reclamation Klamath Project daily operational accounting (2024 and 2026 water years).

Every work cited in the text is linked above, except the subscription-only Smith et al. (2002) journal article.

What changed since the June 5 post

  • New The early-June mortality event is now quantified: an estimated ~2,170 dead or dying of ~3,500 Chinook caught at the Iron Gate and I-5 traps June 2–5 — nearly all hatchery fish.
  • New A river-mile disease-gradient chart from the upper river down to Weitchpec, with the migration-distance research (Muir/Smith/Williams; Lestelle 2007) and the hatchery-siting and coho-competition point (FERC 2022).
  • New Through June 12 the mainstem catch was a late hatchery pulse (ad-clipped Chinook peaking June 2–5), and fork lengths confirm the wild run is in its tail.
  • Fix The second chart’s net-inflow note is corrected: the A Canal is already accounted for in net inflow, not subtracted from it — the figure is storage change plus the releases, not minus.
  • Unchanged The season-to-date figures in the original analysis (~580,000 wild Chinook, ~80% emigrated before April 2, ~13% / ~75,000 infected) are not affected by the new week of data. The two timelines below now extend through June 21.
  • New Both timelines updated to June 21 from the WY2026 Reclamation accounting and the Karuk Final temperature record: mainstem temperature has climbed back above 24°C, and stored water now covers the prescribed flow on 57 of the 77 days since the April 5 peak.
  • Fix Hatchery location corrected: Fall Creek Hatchery is ~7 miles above the former Iron Gate Dam (~200 river miles from the sea); it released 675,000 Chinook smolts on May 14 (CDFW).
  • New A plain-language summary and a “flows” explainer for non-specialists, plus the dead-fish accounting: ~2,170 documented dead (a floor), ~0.3% of the May 14 hatchery release, ~0% of the wild run.
  • New Three up-front facts (coho absence, the hatchery timing/location link, and the Southern Resident killer whale prey connection), and an “open question” section red-teaming the assumption that more flow always means less disease (anchored to Ray, Holt & Bartholomew 2012).
  • Fix Data attribution clarified per the Karuk Tribe: water temperature is credited to the Tribe’s measured Iron Gate record (used with permission); the flows chart uses Reclamation accounting, not USGS-via-Karuk.
  • New A “What CDFW reported” box: the agency’s release-timing rationale, the second confirmed parasite (P. minibicornis), the natural >99% juvenile mortality context, and CDFW’s note that post–dam-removal C. shasta mortalities have run lower.
  • New The bottom chart now overlays the 2024 (dam-removal) Keno release on the same Oct–Jun calendar, for a year-over-year flow comparison (2024 ReclamationTracker, col I).
  • New The bottom chart also overlays water year 1992 — the drought of record (USGS Keno gage 11509500), which ran below ~670 cfs all year and ~150–200 cfs in spring/June — well below 2026, to test the “low flows” characterization against the measured record.
  • New The “low flows” box now acknowledges the common ground in recent coverage — this is a warm-water-and-parasite event — before drawing the distinction: the operative variable is temperature, 2026 was not a low-flow year, and the demanded remedy is warm lake water.