You do not need a date for the next big earthquake to secure a bookshelf. That distinction matters on the U.S. West Coast, where evidence for dangerous earthquakes is strong but reliable prediction of a particular event remains out of reach. Preparing now makes sense without claiming that a magnitude-8-or-larger earthquake will strike within two years. [1] [2] [3] [6]
The U.S. Geological Survey says there is no reliable method for predicting a major earthquake’s time, location and magnitude. Earthquake swarms, changes in groundwater and unusual animal behavior can occur without a major earthquake following. A possible precursor is not a dependable prediction—and the evidence reviewed does not justify singling out a two-year window. [1]
A long record, not a ticking clock
Cascadia offers compelling evidence of what the region can experience. Buried coastal landscapes, tsunami deposits, offshore sediments, tree rings and Japanese tsunami records together place its last great earthquake on January 26, 1700. Its estimated magnitude was 8.7–9.2, according to a 2025 synthesis by USGS and University of Washington researchers. [2]
That synthesis estimates approximately a 10–15 percent chance of a roughly magnitude-9 earthquake rupturing the full Cascadia margin over 50 years, depending on the recurrence model. For southern Cascadia, it estimates about a 30 percent chance of a magnitude-8-or-larger earthquake over 50 years, including either a partial- or full-margin rupture. [2]
These are model-based probabilities, not measured countdowns. A 50-year estimate does not mean an earthquake will wait five decades; it describes the chance of an event occurring within that interval. USGS uses that horizon partly because it is comparable to a building’s lifetime. The report also identifies uncertainties in past rupture sizes and in applying time-dependent models to southern Cascadia. Neither probability establishes a two-year deadline. [2]
California’s forecast has a starting date
California also has an established hazard from very large earthquakes. The UCERF3 forecast, published in 2015, estimated about a 7 percent chance of a magnitude-8-or-larger earthquake in its Greater California region over the 30 years beginning in 2014. That window extends approximately to 2044; it is not a newly issued forecast for 2026–2056. Its tabulated results exclude Cascadia earthquakes. [3]
UCERF3 incorporated geological fault information and a broader range of possible ruptures, including earthquakes spanning multiple faults. Allowing such ruptures helped increase the estimated likelihood of very large events compared with the preceding model. But the forecast remains an approximation of a complex natural system, not evidence for a particular earthquake date. [3]
Reading the past has uncertainties
Even the ancient-earthquake record requires interpretation. Nora Nieminski and colleagues compared magnetic-susceptibility records—measurements of sediment’s response to a magnetic field—from sediment cores against randomized sequences. Only a small number of core pairs correlated more confidently than the randomized sequences, challenging confidence in some long-distance sediment correlations used to reconstruct earthquake histories. [4]
A separate study by Jenna Hill and colleagues, published in January 2026, used detailed seafloor mapping, subsurface profiles and sediment cores at a southern Cascadia site. It documented how underwater landslides produce deposits corresponding to earthquake-history records, supporting the usefulness of those deposits. [5]
These findings address different aspects of the evidence; they do not establish that every sediment correlation is correct or forecast the next earthquake. Both summaries here draw on abstracts rather than complete papers, limiting the assessment of their methods and conclusions. [4] [5]
Prepare for uncertainty, not a deadline
The practical response is straightforward. CDC recommends securing furniture, appliances and other objects that could fall; preparing emergency water, food, first-aid supplies and important documents; practicing Drop, Cover and Hold On; and making household evacuation and reunification plans. Earthquakes below magnitude 8 can also be dangerous, so that number should not become the threshold for concern. [6]
The strongest reason to act is not that science has identified a looming deadline. It is that the hazard is established while the timing remains uncertain. Preparedness does not require knowing which day an earthquake will happen. [1] [2] [3] [6]
Sources
- [1] Can you predict earthquakes? | U.S. Geological Survey
- [2] Earthquake Probabilities and Hazards in the U.S. Pacific Northwest
- [3]
- [4] Turbidite correlation for paleoseismology | U.S. Geological Survey
- [5] Widespread abyssal turbidites record megathrust earthquake-triggered landslides and coseismic deformation in the Cascadia subduction zone | U.S. Geological Survey
- [6] Preparing for Earthquakes | Earthquakes | CDC