What Is a Foreshock? Warning Signs Before Major Earthquakes

Understanding the Earthquake Sequence

Earthquakes rarely occur in isolation. When stress builds up along a fault and finally releases, it often happens in a sequence of events rather than a single rupture. Scientists categorize these events into three types: foreshocks, the mainshock, and aftershocks.

The mainshock is the largest earthquake in any sequence. It's the event that releases the most energy and typically causes the most damage. As we discuss in our guide to earthquake magnitude scales, scientists measure this energy release to determine the earthquake's size.

Aftershocks are smaller earthquakes that follow the mainshock, occurring in the same general area as fault adjustments continue. We cover these extensively in our article on earthquakes versus aftershocks.

Foreshocks are the precursors—smaller earthquakes that occur before the mainshock. But understanding foreshocks isn't as straightforward as it might seem, and that's where things get complicated.

The Foreshock Problem: Only Hindsight is 20/20

Here's the fundamental challenge with foreshocks: they're only identifiable in retrospect. When a magnitude 4.5 earthquake strikes, there's no way to know whether it's a foreshock warning of a larger quake to come, or just an ordinary earthquake that will be followed by aftershocks.

Imagine you're a seismologist monitoring earthquake activity. A magnitude 5.0 earthquake occurs on a known fault. Is this a foreshock preceding a magnitude 7.0 earthquake in the next few hours or days? An independent earthquake that happens to be near the maximum size for this particular fault segment? Or the mainshock of a sequence that will be followed by smaller aftershocks?

The honest answer is: you simply don't know. And this uncertainty makes foreshocks frustratingly unreliable as warning signs, despite what disaster movies might suggest.

Famous Foreshock Examples (Identified After the Fact)

While we can't use foreshocks to predict future earthquakes, studying them after major events helps scientists understand earthquake sequences better. Here are some notable examples where foreshocks preceded major earthquakes.

Why Don't Most Earthquakes Have Foreshocks?

Statistical studies show that only about 40% of large earthquakes are preceded by foreshocks. The remaining 60% occur with no smaller precursor events. But even that 40% figure is somewhat misleading.

The percentage depends on how you define "foreshock" in terms of time window and distance. If you look at earthquakes within hours and very close to the eventual mainshock location, the percentage drops significantly. Expand the window to days or weeks and include a larger geographic area, and more earthquakes qualify as foreshocks, but they become less useful as warnings.

Furthermore, regions with high background seismicity—places where small earthquakes happen frequently—experience small quakes constantly. California, for example, has thousands of detectable earthquakes every year. The vast majority of these are not foreshocks to anything larger.

What Causes Foreshocks?

Scientists have developed several theories about what causes foreshock sequences, all related to how stress accumulates and releases along faults. Understanding what causes earthquakes helps explain these mechanisms.

The Cascade Model

One theory suggests that any earthquake could potentially trigger a larger one through a cascade effect. When a small earthquake ruptures part of a fault, it redistributes stress to adjacent areas. If one of these stressed areas is already close to failure, it might rupture, creating a larger earthquake. Whether the first event becomes a foreshock or remains the mainshock depends on local conditions and how much stress was already stored in nearby fault segments.

The Nucleation Model

Another theory proposes that foreshocks result from a slow, gradual preparation process before the main rupture. In this model, small patches of the fault begin failing days or hours before the major rupture, creating foreshocks. The fault is essentially warming up before the big event. However, this preparation process apparently doesn't happen before all large earthquakes, which is why many occur without foreshocks.

The Preslip Model

Some scientists suggest that foreshocks result from slow, creeping movement (called aseismic slip) on parts of the fault that then triggers brittle failure creating small earthquakes. This creeping movement gradually loads more stress onto the locked portions of the fault until they catastrophically fail in the mainshock.

The challenge is that these processes can also occur without leading to a major earthquake. The fault might creep a little, produce some small earthquakes, and then stop. Or the redistribution of stress from a small earthquake might not trigger additional, larger ruptures.

Can Animals Predict Earthquakes Based on Foreshocks?

You've probably heard stories about animals behaving strangely before earthquakes—dogs barking excessively, birds flying erratically, or fish jumping out of water. Could animals be detecting foreshocks or other precursor signs that humans miss?

While anecdotal reports abound, scientific evidence remains inconclusive. Some animals might detect P-waves that humans don't notice, effectively feeling small foreshocks before people do. However, animals also behave strangely for countless other reasons unrelated to earthquakes.

The fundamental problem remains: even if animals could reliably detect foreshocks, foreshocks themselves aren't reliable predictors of larger earthquakes. Most foreshock-like earthquakes aren't followed by larger events. For more on this topic, see our article on whether animals can predict earthquakes.

The Statistics Problem

Here's a thought experiment that illustrates why foreshocks can't be used for prediction. Southern California experiences roughly 10,000 earthquakes per year, most too small to feel. Let's say 100 of those are magnitude 4.0 or larger—noticeable earthquakes that might be foreshocks.

Now, suppose one magnitude 7.0 earthquake occurs in California during that year (which would be unusual, but let's use it for illustration). If we treat every magnitude 4.0+ earthquake as a potential foreshock and issue warnings, we'd have 99 false alarms for every real prediction. And that's with very generous assumptions.

In reality, the false alarm rate would be much higher because many smaller earthquakes would also need to be considered. The economic cost and loss of public trust from constant false alarms would be enormous. When a real prediction came, people would likely ignore it because of "warning fatigue."

What About Earthquake Swarms?

Sometimes regions experience earthquake swarms—clusters of many small to moderate earthquakes happening in a short time period, without a single clear mainshock. Are these different from foreshock sequences?

Earthquake swarms occur when stress migrates through a region, often associated with fluid movement or volcanic activity. Unlike a typical mainshock-aftershock sequence where one large earthquake dominates, swarms consist of many similarly-sized events.

The challenge is distinguishing whether a swarm will continue as a swarm, evolve into a foreshock sequence leading to a large mainshock, or simply peter out. Iceland experiences frequent swarms, particularly near volcanoes. The Salton Sea region in Southern California also experiences regular swarms. Most of these swarms don't lead to major earthquakes.

When a swarm occurs, seismologists increase monitoring and inform local officials, but they cannot predict whether something larger will follow.

Modern Monitoring and Early Warning

While scientists can't predict earthquakes using foreshocks, modern technology does provide some benefits when increased seismic activity occurs.

When an earthquake sequence begins, seismographs immediately detect it, and monitoring agencies can assess whether activity is increasing. If a swarm or potential foreshock sequence emerges, scientists can deploy additional temporary seismometers to improve monitoring of the region.

Early warning systems can provide seconds to minutes of warning after a large earthquake begins, giving people time to take protective action before the strongest shaking arrives. These systems detect the initial P-waves from an earthquake and use them to estimate the earthquake's size and predict when stronger S-waves and surface waves will arrive at distant locations.

This isn't prediction—the earthquake has already started—but it does provide life-saving warning time. Japan's system gave Tokyo about 60 seconds of warning before strong shaking arrived during the 2011 Tohoku earthquake, despite the city being 370 kilometers from the epicenter.

The Future of Earthquake Forecasting

While earthquake prediction remains impossible, scientists are working on probabilistic earthquake forecasting. This approach doesn't predict when an earthquake will occur but rather estimates the probability of an earthquake within a certain time period.

For example, after a magnitude 6.0 earthquake, scientists can calculate the probability that a larger earthquake will follow within the next week. These probabilities are typically low—usually a few percent—but they're higher than the background probability before the first earthquake occurred.

Some regions now use operational earthquake forecasting systems that update probabilities in real-time as new earthquakes occur. After a potentially threatening earthquake, officials might take precautionary measures like inspecting critical infrastructure or pre-positioning emergency supplies, without issuing full evacuations that would be based on deterministic predictions.

This statistical approach acknowledges the reality: we cannot predict earthquakes, but we can assess changing probabilities and adjust our preparedness accordingly.

The Bottom Line

Foreshocks are real phenomena—smaller earthquakes that precede larger ones in about 40% of major earthquake sequences. However, they're only identifiable as foreshocks in hindsight, after the larger earthquake has already occurred.

When a small earthquake strikes, there's no reliable scientific method to determine whether it's a foreshock warning of a larger earthquake to come, an independent earthquake that will be followed by smaller aftershocks, or part of a swarm that will dissipate without a major event. This fundamental uncertainty, combined with the high rate of background seismicity in active regions, makes earthquake prediction impossible.

The idea that we might one day predict earthquakes remains appealing, but decades of intensive research have consistently reinforced the same conclusion: the Earth's crust is too complex, and earthquake physics too chaotic, for reliable short-term prediction based on foreshocks or any other precursor phenomena.

Instead of prediction, scientists focus on forecasting long-term probabilities, improving building codes in seismically active regions, developing early warning systems that provide seconds to minutes of warning after an earthquake begins, and educating the public about earthquake preparedness. These approaches, while less dramatic than prediction, save lives and reduce earthquake impacts.

Understanding that foreshocks can't predict earthquakes isn't a failure of science—it's an honest acknowledgment of the limits of our knowledge and the complex, chaotic nature of earthquake processes. The best protection comes not from trying to predict unpredictable events, but from being prepared for earthquakes whenever and wherever they strike.