Satellite images taken on May 22, 2026, and published by the NASA Earth Observatory on June 18, 2026, tell the story in two frames: on the left, June 2023, the San Carlos Reservoir holds roughly 60% of its capacity, appearing wide, blue, and expansive against the Arizona desert. On the right, May 22, 2026: just 389 acre-feet of water remain — less than 1% of the reservoir’s full capacity of approximately 900,000 acre-feet. Where water once stretched across 158 miles of shoreline, the exposed bed of the Gila River is now visible, flanked by green tamarisk and cottonwood colonizing the newly exposed lakebed.
As ScienceDaily reported on June 18, 2026, the reservoir’s collapse triggered a catastrophic hypoxic event — a fatal drop in dissolved oxygen — that killed virtually all aquatic life. On June 5, 2026, the Gila Herald confirmed, the San Carlos Recreation and Wildlife Department closed the lake indefinitely following what officials described as a “catastrophic 100% fish kill.”
KTAR reported: “Officials said the dead and decomposing fish created a public health concern. Visitors are being asked to avoid the affected area while conditions are monitored.”
What Caused the 100% Fish Kill — Converging Cascades in a Drought-Depleted System
The mechanism behind the fish kill is a textbook hypoxic collapse — but the specific combination of factors that triggered it at San Carlos in 2026 reflects the compound vulnerability of a system already at its breaking point. As the Gila Herald’s investigation detailed, several volatile factors converged:
Historic winter snowpack failure. The Gila River’s primary water source is winter snowpack accumulation in the Mogollon Mountains and Black Range of southwestern New Mexico. The 2025–2026 winter produced near-zero snowpack in this watershed — one of the worst on record — leaving the spring runoff to San Carlos essentially nonexistent.
Downstream irrigation releases from Coolidge Dam. Part of the blame for the accelerated depletion was attributed to downstream irrigation demands in agricultural communities, forcing the release of stored water from Coolidge Dam. The dam, constructed between 1924 and 1928, provides irrigation water for the San Carlos Irrigation Project — meaning the reservoir’s supply competes directly with agricultural demand during drought.
Thermal-oxygen death spiral. Warm water holds significantly less dissolved oxygen than cold water. As the lake dropped toward a fraction of its capacity, the remaining shallow pools heated rapidly under the Arizona desert sun, reducing oxygen availability at the same time that millions of fish were compressed into increasingly tight, stagnant pockets — consuming the diminishing oxygen supply faster than it could be replenished. An algal bloom added the final blow: in April 2026, aerial photos showed characteristic surface scum across the lake, and overnight oxygen draw from the bloom drove dissolved oxygen below the survival threshold for fish in a matter of hours.
The result was total: as Gizmodo reported on June 8, 2026, San Carlos Lake lost “approximately 100% of the fish population,” according to official authorities. Species living in the reservoir included largemouth bass, black crappie, yellow perch, channel catfish, and the lake’s trophy fish — flathead catfish, one of which produced a nearly 72-pound individual described as one of Arizona’s largest recorded.
San Carlos Lake Collapse — Key Data (June 2026)
Detail
Reservoir capacity (full)
~900,000 acre-feet
Water remaining May 22, 2026
389 acre-feet (less than 1% of capacity)
NASA Earth Observatory coverage
Published June 18, 2026
Lake closure date
June 5, 2026 (indefinite)
Fish mortality
Approximately 100% of all fish species
Cause
Hypoxia (oxygen depletion) from drought + thermal stress + algal bloom
Contributing factor
Downstream irrigation releases from Coolidge Dam
Shoreline when full
158 miles
Annual visitors (normal years)
~200,000
Located on
San Carlos Apache Reservation, ~100 miles east of Phoenix
Primary function of reservoir
Agricultural irrigation for San Carlos Irrigation Project
Recent parallel event
Cluff Pond #3 (Graham County) — separate fish kill, same mechanism, same weeks
Prior kills at San Carlos
2018, 2022, and historical events (1976–77 killed estimated 5 million fish)
Prior recovery timeline (1977 event)
Five-year recovery period
Current restrictions
No fishing, no fish harvesting, no recreational access to affected area
The Public Health Dimension — What a 100% Fish Kill Does to a Community
Fish kills at the scale documented at San Carlos Lake do not affect only wildlife. As the San Carlos Recreation and Wildlife Department explicitly warned in their closure notice, as reported by KTAR: “decomposing fish may pose health risks to individuals who enter the area or attempt to fish.”
The decomposition of millions of fish produces several biological hazards:
Hydrogen sulfide and ammonia gas. Anaerobic decomposition of organic matter — including fish tissue — produces hydrogen sulfide (the odor compound responsible for the “rotten egg” smell common near stagnant bodies of water) and ammonia. At the concentrations produced by large-scale fish decomposition in a shallow, enclosed basin in desert heat, these gases can cause respiratory irritation in individuals near the shoreline, with more severe effects possible for people with asthma, COPD, or other respiratory conditions.
Bacterial blooms. Decomposing organic matter and the hypoxic conditions that enabled the fish kill both promote explosive growth of anaerobic bacteria. Some bacterial species that thrive in these conditions are human pathogens. Contact with water containing high bacterial loads — through swimming, wading, or incidental exposure during fishing — carries infection risk. The public closure specifically bars all shoreline recreation to prevent this exposure.
Downstream water quality. San Carlos Lake primarily serves as an agricultural water source for the San Carlos Irrigation Project. As the water in the reservoir becomes more concentrated with decomposition byproducts, bacterial metabolites, and reduced dissolved oxygen, the quality of water available for agricultural use downstream declines. The full downstream implications for crops and water quality depend on how much water remains usable and how quickly decomposition byproducts move through the system.
Subsistence and food security. For members of the San Carlos Apache Tribe who rely on the lake for subsistence fishing, the total loss of the fish population eliminates a traditional food source. Daniel Juan, senior biologist for the San Carlos Recreation and Wildlife Department, told the Globe-Miami Times that the lake will require a significant recovery period before it can be restocked — and that recovery depends entirely on returning reservoir levels, which are governed by precipitation events the community cannot control.
Can San Carlos Recover — and What Recovery Requires
The Gila Herald reported that the reservoir has nearly emptied at least 20 times since Coolidge Dam’s construction and reached full capacity on only three occasions in nearly 100 years. Following the 1976–1977 drought event, the ecosystem required a five-year recovery period.
Recovery requires two things: enough precipitation to refill the reservoir to levels that support dissolved oxygen above survival thresholds, and subsequent restocking by state and tribal wildlife officials once those conditions are established. The National Weather Service’s 2026 Arizona monsoon outlook offers some hope — the monsoon season, which typically begins in July, can deliver significant precipitation to the Gila River watershed. ScienceDaily’s June 18 coverage noted that “heavy summer rains could help the reservoir rebound.” But given the catastrophic baseline — less than 1% of capacity and complete ecosystem loss — a single monsoon season is unlikely to restore the fishery.
The broader Arizona reservoir system remains under compound stress as of June 2026: the Colorado River basin snowpack ran below average across the 2025–26 winter season, Lake Mead closed the year at levels triggering operational restrictions on downstream water allocations, and surface temperatures across the state’s lower-elevation lakes exceed historical June averages.
Frequently Asked Questions
What happened at San Carlos Lake in 2026?
A historic lack of winter snowpack in the Gila River watershed caused San Carlos Reservoir to drop below 1% of its full capacity by May 2026. Combined with heat and algal bloom, this produced hypoxia — a fatal oxygen depletion — that killed approximately 100% of the lake’s fish population. Officials closed the lake indefinitely on June 5, 2026.
What are the public health risks from the fish kill?
Decomposing fish produce hydrogen sulfide gas and ammonia, which can cause respiratory irritation near the shoreline. Bacterial blooms in hypoxic water create pathogen exposure risk for anyone entering the water. Downstream agricultural water quality may also be affected. The San Carlos Recreation and Wildlife Department warned that decomposing fish may pose health risks and has banned all shoreline recreation.
How full was the reservoir?
NASA Landsat satellite imagery captured on May 22, 2026, showed just 389 acre-feet remaining — less than 1% of the reservoir’s full capacity of approximately 900,000 acre-feet. The NASA Earth Observatory published these images on June 18, 2026.
Has this happened before at San Carlos?
Yes. The reservoir has nearly emptied at least 20 times since the Coolidge Dam was constructed in 1930. A catastrophic fish kill in 1976–77 killed an estimated 5 million fish and required a five-year ecosystem recovery. Fish kills also occurred in 2018 and 2022.
When can the lake reopen?
The lake is closed indefinitely. Recovery requires sufficient monsoon-season precipitation to raise water levels and dissolved oxygen, followed by wildlife agency restocking. Given the complete ecosystem loss and the severity of the drought, recovery to pre-event conditions will likely require multiple seasons.
