- Study Links Major Emitters to Western Water Loss
- Researchers linked major industrial emitters to less Western snowpack and streamflow.
- The study estimated a 15 percent decline in April snowpack tied to those emissions.
- Warm-season streamflow fell an estimated 6 percent, while irrigation demand rose 2.4 percent.
- The Upper Colorado River Basin showed declines in both cool- and warm-season streamflow.
- Researchers also linked the emissions to part of California’s Central Valley groundwater loss.
Thursday, August 27, 2026 — A new study has tried to answer a difficult question about the West’s shrinking water supply: How much of the change can be traced to emissions from the world’s largest fossil fuel and cement producers?
The researchers found a measurable connection.
The study, published in Communications Earth & Environment
, examined water conditions across the continental United States west of 105°W. Researchers studied snowpack, streamflow, irrigation demand, and groundwater.
They estimated that emissions since 1950 traced to a group known as the “Carbon Majors” accounted for roughly 40 to 64 percent of the changes in those water measurements that the study attributed to human-caused climate change.
The findings do not mean those companies caused all of the West’s water problems. Water supplies are also affected by natural climate variability, long-standing water overallocation and other factors. In California’s Central Valley, for example, the researchers treated about 70 percent of groundwater depletion as resulting from overallocation and about 30 percent as related to climate factors.
What the study attempts to do is separate one part of a complicated water problem and estimate how large that part may be.
Who Are the “Carbon Majors”?
The term “Carbon Majors” refers to 122 major oil, natural gas, coal and cement-producing entities included in the database used by the researchers.
According to the study, roughly 70 percent of global industrial carbon dioxide emissions since 1854 can be traced to those producers and cement manufacturers. About 97 percent of those emissions occurred after 1950.
The researchers focused much of their analysis on emissions since 1950.
Rather than simply looking at how much the climate has warmed, they used climate and water models to create versions of the past in which certain emissions were removed. They then compared those modeled conditions with what actually occurred.
In simpler terms, the researchers asked what Western water conditions might have looked like without the warming associated with those emissions.
Less Snow Stored in the Mountains.
Snow is one of the West’s most important natural water-storage systems.
Winter storms leave snow in the mountains. That snow can remain there for months before melting and feeding rivers and streams during spring and summer, when farms and communities need large amounts of water.
The researchers examined the amount of water contained in the snowpack on April 1, a date near the seasonal snowpack peak in many Western basins.
For the decade from 2014 through 2024, they estimated that emissions traced to the Carbon Majors since 1950 were associated with a 15 percent reduction in April 1 snow water across the West.
The researchers calculated that reduction at about 28 million acre-feet per year. They described that amount as roughly equal to the maximum storage capacity of Lake Mead.
For comparison, the study estimated that all human-caused climate change produced a 36 percent reduction in April 1 snow water. The Carbon Majors portion represented about 42 percent of that estimated climate-related decline.
The largest total snowpack reductions attributed to the Carbon Majors occurred in California and the Pacific Northwest.
Less Water Arriving During the Warm Season.
The study found another important change downstream from the snowpack.
Across the West, researchers estimated that emissions traced to the Carbon Majors since 1950 were associated with a 6 percent reduction in warm-season streamflow from April through September during the 2014 through 2024 period.
That amounted to about 12 million acre-feet of streamflow per year.
The study estimated that all human-caused climate change reduced warm-season streamflow by about 13 percent. The Carbon Majors contribution represented about 47 percent of that climate-related reduction.
Timing changed as well.
Researchers estimated that the center point of runoff timing shifted an average of about 5.6 days earlier across the West because of emissions traced to the Carbon Majors. In some mountainous areas, including the western slopes of Washington’s Cascade Mountains, the shift reached as much as 30 days.
That matters because water arriving earlier is not necessarily available when demand is highest later in the year.
The Colorado River Stands Out.
The changes estimated for the Colorado River Basin were smaller than those found in California and the Pacific Northwest, but the researchers said even relatively small changes can matter in a river system where water is already overallocated.
The Upper Colorado River Basin was also different from most of the other basins examined.
In most areas, the researchers found increased streamflow during the cooler months along with decreased flow during the warmer months. In the Upper Colorado River Basin, however, the study found decreases during both periods.
Across the Upper and Lower Colorado River basins combined, researchers attributed an estimated 3 percent decline in streamflow, about 0.43 cubic kilometers per year, to Carbon Majors emissions since 1950. The study noted that relatively small changes can have “outsized effects” on the Colorado River’s water budget because the river is overallocated.
Crops Need More Water at the Same Time.
The study found pressure on the other side of the water equation as well.
While the available supply was declining, the amount of water needed for irrigation was increasing.
Researchers estimated that emissions traced to the Carbon Majors since 1950 increased annual irrigation demand across the West by about 2.4 percent during the 2014 through 2024 period. That represents approximately 700,000 acre-feet of additional irrigation demand each year.
All human-caused climate change was associated with an estimated 4.3 percent increase.
The researchers calculated that the Carbon Majors contribution represented about 57 percent of the climate-related increase in irrigation demand.
The problem is partly one of timing.
Irrigation demand is highest during the warm months, when streamflow is generally lowest. Higher temperatures increase the amount of water that can evaporate from the landscape and vegetation. That can increase the amount of water crops require even if the amount of farmland does not increase.
The study found some of the largest increases in irrigation demand in central Washington, southern Idaho and California’s Central Valley. The strongest relative increases occurred in the Pacific Northwest and Upper Colorado River Basin.
California Groundwater Shows Another Effect.
The researchers also took a closer look at California’s Central Valley, where groundwater has been heavily used for agriculture.
From 2003 through 2024, the study reported an observed groundwater decline of 33.9 cubic kilometers.
The researchers assumed about 70 percent of that decline resulted from long-term groundwater overallocation and about 30 percent resulted from climate-related factors, including human-caused climate change and natural variability. That left about 10.2 cubic kilometers of observed groundwater loss associated with climate factors.
Their modeling attributed about 6.5 cubic kilometers of groundwater decline to human-caused climate change. Of that, approximately 3.5 cubic kilometers was attributed to emissions traced to the Carbon Majors since 1950.
That means the researchers attributed roughly 34 percent of the Central Valley’s climate-related groundwater loss from 2003 through 2024 to those emissions. It represented roughly 10 percent of the total observed groundwater loss during that period.
The study’s groundwater calculation is especially important to read carefully. The researchers did not attribute one-third of all Central Valley groundwater depletion to the Carbon Majors. Their estimate was about one-third of the portion considered climate-driven.
Why the Researchers Tested Two Methods.
Part of the study addressed a question that may eventually matter beyond scientific journals.
Researchers compared two ways of calculating responsibility for climate-related water changes.
One is a detailed “attribution” method that models the chain from emissions to warming and then to changes in snowpack, streamflow, irrigation demand and groundwater.
The other is a simpler proportional method. That approach assumes that if an emitter is responsible for a certain share of warming, it can be assigned the same proportion of a particular climate impact.
The study found that the simpler method did not always work equally well.
Across the entire study area, the proportional approach underestimated increased irrigation demand by 19 percent compared with the more detailed attribution method. It overestimated reductions in April snow water by 13 percent and warm-season streamflow by 4 percent.
The researchers said the proportional method may be useful for some purposes but could be less suitable when calculating damages, where greater precision may be necessary.
The Study Has Limits.
The authors also described uncertainties in their work.
Their estimates depend on climate models, choices about model settings and hydrologic modeling. The researchers said their uncertainty calculations may not capture the full range of possible structural uncertainty.
The Central Valley groundwater analysis used a relatively simple linear model. It also did not directly account for water-management decisions such as managed aquifer recharge or changes in irrigation practices.
The researchers therefore treated 30 percent of recent Central Valley groundwater depletion as climate-related, which they described as a conservative estimate. They said future work could use more sophisticated groundwater modeling to improve the estimates.
Those limitations are significant because the study is attempting something unusually complicated: tracing emissions through the climate system and then through the water cycle to individual changes in Western water supplies and demand.
Science That Could Reach the Courtroom.
The research also has a legal and policy dimension.
States and local governments have pursued lawsuits and other policies involving fossil fuel producers. Those efforts raise a difficult question: Even when climate change contributes to a loss, how much of that particular loss can reasonably be connected to emissions associated with specific producers?
The researchers said their findings could provide scientific information for climate liability cases, climate-related funding laws, economic damage calculations and future liability frameworks.
That does not resolve those legal questions. The study itself notes that science, water management, public policy and law may require different levels of precision when using attribution estimates.
What the research provides is a new attempt to put numbers on one part of the West’s water problem.
The resulting picture is not simply one of less water.
According to the study, warming associated with emissions traced to major fossil fuel and cement producers has contributed to less mountain snow, less streamflow during the warm season, earlier runoff and greater irrigation demand. In some places, those changes also increase pressure on groundwater.
For the Colorado River Basin, the estimated changes are smaller than in some other parts of the West. But the researchers point to the river’s existing overallocation as the reason even relatively small changes in supply and demand can have large consequences for its water budget.
Citation.
Williams, Emily L., John T. Abatzoglou, Carly A. Phillips, Mohammad Safeeq, Shaina Sadai, Oriana S. Chegwidden, Angel Santiago Fernandez-Bou, L. Delta Merner and J. Pablo Ortiz-Partida. “Carbon emissions exacerbating the Western US water crisis.” Communications Earth & Environment, 7, 675 (2026). DOI: 10.1038/s43247-026-03900-6. The manuscript was accepted July 15, 2026.
