Yes. Tokyo has earthquakes constantly, and the city sits on one of the most seismically active pieces of real estate on Earth. The greater Tokyo region records thousands of felt and unfelt tremors every year, and the metropolis is built directly over the boundary where four tectonic plates converge. A typical resident experiences several noticeable shakes annually, and the Japanese Meteorological Agency (JMA) issues frequent intensity reports for the capital. The more useful question is not whether Tokyo has earthquakes, but which faults produce them, how often they turn violent, and what that means for anyone living there or visiting.

Four plates under one city

Tokyo's seismicity starts with plate geometry. The Pacific Plate dives westward beneath Japan at roughly 8 to 9 centimeters per year. The Philippine Sea Plate rides over it and subducts northward beneath the Kanto region at a comparable rate. The Okhotsk (North American) Plate caps the archipelago from the north. Where these slabs meet beneath the Kanto Plain, they create a stacked sandwich of crust and sinking ocean floor.

Each interface is a fault. The megathrust between the Philippine Sea Plate and the Okhotsk Plate runs close beneath Tokyo Bay and the surrounding lowlands. The deeper boundary between the Philippine Sea and Pacific plates sits below that. When these locked surfaces slip, the rupture releases centuries of accumulated strain in seconds. The shallow Kanto Plain sediments then amplify the shaking, because soft soil resonates with seismic waves the way a bowl of jelly wobbles on a table.

That amplification is why Tokyo's damage patterns are patchy. A tremor that rattles bedrock neighborhoods in western Tokyo can liquefy reclaimed land along the bay. Plain language: the ground under different parts of the city behaves like different materials, so the same earthquake hits them unequally.

How often, how hard

Japan experiences roughly 1,500 felt earthquakes per year nationwide, according to JMA monitoring, though the count varies. Tokyo's share is substantial. Since the 2011 Tohoku earthquake, the Kanto region has recorded a pronounced increase in small and moderate events, a pattern seismologists call induced seismicity—stress changes from the massive offshore rupture nudged nearby faults closer to failure.

The JMA shindo scale measures shaking intensity at a location rather than magnitude at the source. It runs from 0 (imperceptible) to 7, split into lower and upper tiers at 5 and 6. Most Tokyo tremors register shindo 1 to 3: felt, rattling, rarely damaging. Shindo 5-lower is where unsecured furniture starts moving and older buildings crack. Shindo 6-upper and 7 cause widespread structural failure. The 2011 Tohoku quake produced shindo 5-upper in parts of Tokyo, more than 200 miles from the epicenter, entirely through long-period wave amplification.

Magnitude and intensity are not the same thing. A magnitude 5.5 directly under the city can produce stronger local shaking than a magnitude 9 offshore. Distance, depth, and geology decide what a resident actually feels.

Timeline: what Tokyo's earthquake history shows

Year Event Approximate magnitude Effect on Tokyo
1703 Genroku earthquake ~8.0 Severe damage, thousands of deaths in the Kanto region
1855 Ansei Edo earthquake ~7.0 Killed an estimated 7,000-10,000 in Edo (Tokyo)
1923 Great Kanto earthquake ~7.9 About 105,000 deaths; firestorms destroyed much of the city
1987 Chiba earthquake ~6.7 Moderate damage, notable for occurring in a previously quiet zone
2011 Tohoku earthquake ~9.0 Shindo 5-upper in Tokyo; triggered aftershock surge across Kanto
2005-present Ongoing background seismicity M4-M6 range Multiple felt events annually; frequent JMA advisories

The 1923 rupture is the reference disaster. It struck at lunchtime, when cooking fires were lit, and the resulting firestorms killed far more people than the collapse of buildings. That detail shaped modern Tokyo's earthquake doctrine: fire, not just shaking, is the primary killer.

Edge cases that change the answer

Not every Tokyo earthquake comes from the same mechanism, and the distinctions matter.

Shallow crustal faults run through and near the city. The 1987 Chiba event surprised seismologists by rupturing a fault that had been considered low-risk. Megathrust events on the Philippine Sea Plate boundary can strike close to the city, producing intense local shaking with little warning time. Deep intraslab earthquakes occur 60 to 100 kilometers down within the subducting plates. They shake a wide area but arrive with less surface rupture. Long-period ground motion from distant megathrusts sways tall buildings in central Tokyo even when low-rise neighborhoods feel little. Induced swarms follow large offshore quakes, raising the background rate for years. Reclaimed bay-area land liquefies during sustained shaking, sinking buildings and rupturing buried pipes.

The counterintuitive detail: a moderate quake directly under the city is often more dangerous than a great one offshore. Depth and proximity determine local intensity, and Tokyo's most feared scenario is a shallow rupture beneath the bay, not another Tohoku-scale event far away.

Misconceptions worth retiring

The most common error is treating "the big one" as a single known future event. Tokyo's risk comes from many possible ruptures on different faults, each with its own probability estimate. A second misconception is that earthquake prediction is reliable. JMA issues long-term probability forecasts, not dated warnings. The agency's Earthquake Early Warning system detects the fast P-waves and broadcasts an alert seconds before the slower, damaging S-waves arrive. Those seconds are enough to stop elevators, halt trains, and trigger automated shutdowns, but not enough to evacuate a building.

A third misconception: modern Tokyo is invulnerable. Building codes tightened dramatically after 1981, and newer towers perform well, but a large fraction of the housing stock predates those standards. Retrofitting has been slow, and older wooden structures in dense neighborhoods remain the weak point.

What to actually do

Tokyo's preparedness culture is unusually mature, and those practices transfer elsewhere.

Know your building's construction year. Anything permitted before 1981 in Japan meets older, weaker seismic standards. Secure tall furniture, bookcases, and water heaters with brackets and straps; most earthquake injuries come from falling objects, not collapse. Keep a go-bag with water, a flashlight, a portable radio, and a first-aid kit; JMA advisories can disrupt power and transit for days. Identify the nearest designated evacuation area and learn two routes to it, because streets may be blocked. Register for JMA's alert apps and enable emergency notifications on your phone. During shaking, drop, cover, and hold under sturdy furniture; do not run outside, where falling glass and facade panels are the hazard.

The threshold that matters most: if you feel shaking that lasts longer than about 20 seconds, treat it as a potentially damaging event and check for gas leaks and structural cracks afterward rather than assuming it was minor.

Limitations of what we know

Earthquake forecasting carries real uncertainty. Probabilities are expressed over decades, not days, and a 30-year estimate of roughly 70 percent for a major Kanto earthquake reflects modeling assumptions rather than a schedule. Monitoring networks are dense, but the deep plate interfaces beneath the Kanto Plain remain poorly imaged, which limits how precisely the next rupture can be located or sized. Any statement that Tokyo is "due" for a quake should be read as a statement about accumulated strain, not a countdown. What is certain is the mechanism: four plates, locked boundaries, and a city built on soft ground above them. The earthquakes will keep coming, and Tokyo's response has been to engineer, educate, and monitor around that certainty rather than deny it.