Roughly 900 million people live along the edges of the Pacific Ocean where tectonic plates collide, and about 90 percent of the planet’s earthquakes strike in that same narrow band. The zone, known as the Ring of Fire, is a horseshoe-shaped belt of faults, volcanic arcs, and oceanic trenches that stretches from New Zealand north through Japan, across the Aleutian Islands, and down the western coasts of North and South America. The concentration of seismic energy in this single corridor shapes building codes, insurance markets, and emergency planning for dozens of countries, and the question of whether that 90 percent figure still holds up under modern monitoring is more than academic.
Why the 90 percent earthquake share demands fresh scrutiny
The figure itself comes from the U.S. Geological Survey, which states that about 90 percent of the world’s earthquakes occur along the Circum-Pacific belt. A second, smaller seismic corridor, the Alpide belt running from the Mediterranean through the Himalayas, accounts for only 5 to 6 percent. Those two zones together leave less than 5 percent of global seismicity scattered across all other regions.
The 90 percent claim has circulated in textbooks, newsrooms, and government briefings for decades. But the number was established during an era when seismograph coverage outside the Pacific Rim was far thinner. Since 2000, hundreds of new stations have come online in Africa, the Middle East, and interior Asia, capturing smaller earthquakes that older networks missed entirely. That expansion raises a practical question: has improved detection in non-Ring of Fire regions measurably narrowed the gap, at least for moderate and smaller events?
The USGS itself draws a distinction by magnitude. When the agency narrows the lens to the planet’s largest earthquakes, the Circum-Pacific belt’s share drops to about 81 percent. The difference between 90 percent for all earthquakes and 81 percent for the largest ones suggests that the headline figure is sensitive to which events get counted and at what magnitude threshold. For the biggest quakes, the Alpide belt and mid-ocean ridges claim a larger slice than the blanket 90 percent figure implies.
USGS data, the ISC-GEM catalog, and what the numbers actually measure
Two primary evidence bases support the global earthquake distribution picture. The first is the USGS Earthquake Hazards Program, which maintains real-time monitoring and publishes the canonical graphic describing the Ring of Fire as a zone of frequent earthquakes and volcanic eruptions encircling the Pacific Basin. The agency’s explainer pages cite specific events to anchor the claim: the 1960 Chile earthquake at magnitude 9.5, the largest instrumentally recorded quake in history, and the 1964 Alaska earthquake at magnitude 9.2. Both struck squarely within the Circum-Pacific belt.
The second evidence base is the ISC-GEM Global Instrumental Earthquake Catalogue, a peer-reviewed dataset compiled by the International Seismological Centre that spans approximately 110 years from 1900 to 2009. The catalog relocates and homogenizes earthquake magnitudes to a common moment-magnitude scale, making cross-era comparisons more reliable. A companion paper published in Earth and Planetary Science Letters analyzed location and seismicity patterns drawn from that catalog, providing the kind of spatial breakdown needed to test whether 90 percent remains accurate across the full instrumental record.
A separate USGS-authored evaluation examined how complete the ISC-GEM catalog actually is, probing which magnitude thresholds and time periods hold up to scrutiny and which do not. That completeness analysis matters because any global percentage is only as good as the detection network behind it. If stations in the early twentieth century missed small earthquakes outside the Pacific Rim, the 90 percent figure for that era could reflect monitoring gaps as much as geology.
The distinction is not trivial for the hundreds of millions of people living in seismically active zones outside the Ring of Fire. Countries along the Alpide belt, from Turkey and Iran to Nepal, have experienced devastating earthquakes that fall within the remaining 5 to 6 percent. If improved detection reveals that the non-Ring of Fire share is actually higher than legacy estimates suggest, those countries may need to recalibrate risk models and building standards accordingly.
Gaps in the record and what to watch for next
The available USGS pages state the 90 percent figure without publishing the underlying calculation, including the magnitude cutoff, time window, or spatial boundary used to produce it. No table or dataset in the supplied sources shows the exact methodology. That absence makes it difficult to determine whether the number reflects all detected earthquakes above a certain threshold, only those above magnitude 4 or 5, or some other filter.
The ISC-GEM catalog papers describe seismicity patterns and completeness characteristics, but neither the catalog overview nor the peer-reviewed spatial analysis contains a direct statement confirming or updating the 90 percent value. Instead, they emphasize how detection capabilities improved over the twentieth century, particularly after the installation of global digital networks in the 1960s and 1970s. For smaller magnitudes, the catalog is incomplete in the early decades and becomes progressively more representative over time.
That evolution matters because the oft-quoted 90 percent figure implicitly mixes together very different eras of monitoring. A global percentage calculated from events after, say, 1970 would rest on a denser, more uniform network than one spanning the entire twentieth century. Without clarity on the time window and magnitude range, readers are left to assume that the number applies equally well to tiny, moderate, and great earthquakes alike, even though the USGS already distinguishes an 81 percent share for the largest events.
Another complication is how boundaries are drawn. The Ring of Fire is not a single line but a broad swath of interacting plates, subduction zones, and back-arc basins. Depending on where analysts choose to truncate that band-how far inland they extend it in the Americas, or how they treat complex junctions near Indonesia and Papua New Guinea-the share of earthquakes assigned to the belt could shift by several percentage points. The same is true for the Alpide belt, which threads through densely populated regions where station coverage has improved dramatically in recent decades.
For policymakers and the public, the precise percentage may matter less than the underlying message: seismic risk is highly concentrated but not exclusive to the Pacific Rim. The USGS materials underscore that some of the world’s deadliest earthquakes have struck outside the Ring of Fire, including events in the Mediterranean and Himalayan regions. Those quakes fall within the 5 to 6 percent attributed to the Alpide belt, yet their human and economic toll rivals or exceeds that of many Pacific events.
Looking ahead, the most informative updates are likely to come from fresh analyses that explicitly combine modern global catalogs with transparent definitions of the Ring of Fire and Alpide belts. Such work would ideally publish the magnitude thresholds used, the time spans considered, and the geographic polygons defining each zone. With those ingredients, independent teams could reproduce and stress-test any global percentage, rather than treating the 90 percent and 81 percent figures as fixed constants.
In the meantime, the legacy numbers remain useful shorthand, with important caveats. They capture the broad reality that the Pacific margins dominate global seismicity, especially for the largest earthquakes that drive tsunami and long-range shaking hazards. But they should not be read as precise, immutable ratios, nor as evidence that regions outside the Ring of Fire are comparatively safe. As monitoring networks continue to expand and catalogs like ISC-GEM are refined, the exact shares may shift, even if the overall pattern of concentration does not.
For communities from Santiago and Tokyo to Istanbul and Kathmandu, the practical takeaway is the same: local hazard assessments should lean on region-specific data and building standards, rather than on a single global percentage. The Ring of Fire may host most of the world’s earthquakes, but the next catastrophic event-wherever it falls within those percentages-will be judged not by its contribution to a statistic, but by how well societies have prepared for the shaking they know is coming.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
