유감 반경 계산기

Calculate how far an earthquake can be felt and what intensity you would experience.

Calculation

지진 체감 반경이 결정되는 방법

지진의 체감 반경은 진앙으로부터 사람들이 진동을 인지할 수 있는 최대 거리입니다. 이는 주로 규모와 깊이에 따라 달라지지만, 지역 지질, 시간대(가만히 앉아 있을 때 진동을 더 잘 느낌), 건물 유형에도 영향을 받습니다. 지진학적 모델은 수정 메르칼리 진도(MMI)가 거리에 따라 어떻게 감소하는지를 설명하는 진도 감쇠 관계를 사용하여 체감 반경을 추정합니다. 인간이 인지할 수 있는 임계값은 대략 MMI II로, 건물 상층에 있는 사람들이 느낄 수 있는 약한 진동입니다.

깊이는 체감 반경을 결정하는 데 중요한 역할을 합니다. 깊이 5 km의 얕은 지진은 지표면 근처에 에너지를 집중시켜 강렬하지만 국지적인 진동을 생성합니다. 깊이 100 km의 동일 규모 지진은 훨씬 더 넓은 범위에 에너지를 분산시켜, 더 넓은 체감 범위를 가지지만 최대 강도는 낮습니다. 이것이 깊은 지진이 종종 매우 먼 거리에서 감지되는 이유입니다. 2013년 깊이 609 km의 규모 8.3 오호츠크해 지진은 러시아 전역에서 감지되었지만, 먼 경로에 따른 에너지 감쇠로 인해 피해는 미미했습니다.

수정 메르칼리 진도 척도

MMI I~II: 감지되지 않거나 거의 감지되지 않음. 지진계 또는 상층에서 가만히 있는 소수의 사람들만 감지.
MMI III~IV: 실내에서 많은 사람들이 감지. 매달린 물체가 흔들리며 지나가는 트럭과 비슷한 느낌. 피해 없음.
MMI V~VI: 거의 모든 사람이 감지. 식기 파손, 선반에서 책이 떨어짐, 경미한 벽 균열. 약간의 구조적 손상 가능.
MMI VII+: 부실 구조물에 피해. 서 있기 어려움. 잘 설계된 구조물도 MMI VIII 이상에서 중간 정도의 피해 가능.

일반적인 용도

비상 계획 및 대국민 소통을 위해 지진이 느껴지는 범위를 추정합니다.
지진 깊이와 인지 가능한 진동 범위 간의 관계 이해.
위험 평가를 위해 다양한 지진 시나리오의 잠재적 영향 범위 비교.

How to Use

1
Enter Magnitude and Depth
Input the earthquake's moment magnitude (Mw) and focal depth in kilometers. Depth is the most important modifier of felt radius after magnitude—shallow events (< 15 km) are felt over smaller but more intensely shaken areas.
2
Select Regional Attenuation
Choose your tectonic region. Eastern continental regions (eastern US, stable cratonic areas) have lower attenuation and transmit seismic waves farther than western active tectonic regions (western US, Japan) where the crust attenuates energy more rapidly.
3
Read the Intensity Map
View the estimated MMI contours at distances ranging from the epicenter to several hundred kilometers. Each contour corresponds to a descriptor from 'not felt' to 'violent shaking,' based on USGS ShakeMap attenuation relations.

About

The geographical extent of earthquake felt shaking depends on a cascade of physical processes that attenuate seismic energy from the source to distant sites. Ground Motion Prediction Equations (GMPEs) are the mathematical backbone of felt-radius estimation, empirically derived from hundreds of thousands of ground motion recordings worldwide. Modern GMPEs are region-specific: the eastern US crustal model (Atkinson and Boore 2006) predicts felt radii roughly twice those of equivalent-magnitude western US events because the cold, rigid eastern craton transmits seismic waves with lower attenuation (higher Q values) than the warm, fractured western crust.

The USGS 'Did You Feel It?' citizen science platform has transformed intensity estimation. Since its launch in 1997, it has collected over 100 million intensity reports from millions of contributors worldwide, providing dense spatial coverage of felt shaking that instrumental networks alone cannot match. Statistical algorithms convert individual reports—rating shaking intensity on a structured questionnaire covering felt motion, sounds, object movement, and structural effects—into aggregate community intensity values that correlate strongly with instrumental measurements. This crowdsourced intensity data is now incorporated into ShakeMap products alongside seismograph recordings.

Felt radius data over decades reveals systematic regional patterns. In the stable cratonic interior of the eastern US, M5.0 earthquakes have been felt at distances exceeding 1,000 km—the 2011 Virginia M5.8 was felt from Georgia to Nova Scotia. In California's tectonically active crust, a M5.0 is typically felt within 200 km. The New Madrid Seismic Zone, straddling Missouri, Arkansas, and Tennessee, presents a particular challenge: the region has produced M7.0–M8.0 events (estimated, 1811–1812) and sits atop thick, seismically efficient sedimentary sequences; a repeat sequence today would impact a population orders of magnitude larger than the sparse frontier settlements of the early 19th century.

FAQ

수정 메르칼리 진도 척도란 무엇인가요?

The Modified Mercalli Intensity (MMI) scale, developed by Giuseppe Mercalli in 1902 and revised by Harry Wood and Frank Neumann in 1931, describes the intensity of ground shaking at a specific location on a Roman numeral scale from I to XII. MMI I is 'not felt except by a very few under especially favorable conditions.' MMI IV is 'felt indoors by many'; dishes rattle. MMI VI is 'felt by all'; heavy furniture moves, some plaster falls. MMI VIII causes 'considerable damage in ordinary buildings'; partial collapse of poorly built structures. MMI X describes 'most masonry and frame structures destroyed.' MMI XII represents 'total destruction'; objects thrown into the air. Reports from residents after earthquakes are systematically collected by USGS 'Did You Feel It?' and used to map intensity distributions.

Why do some earthquakes feel different even at the same distance?

Several factors cause shaking to feel different at identical distances from the same earthquake. Soil amplification is the dominant local effect: soft sediments amplify ground motion and extend its duration compared to hard rock. Local geology can amplify shaking by factors of 2–10 in peak ground velocity. Building resonance also plays a role: tall buildings sway more during long-period waves that travel farther from the source, so a high-rise in a distant city may sway noticeably while a single-story home nearby is unaffected. The radiation pattern of the earthquake source—controlled by fault geometry—creates directional variations in amplitude. Finally, topographic effects (ridge amplification, valley trapping) can locally increase or decrease shaking by factors of 1.5–2.

How many people typically feel a large earthquake?

USGS PAGER (Prompt Assessment of Global Earthquakes for Response) estimates population exposure to shaking levels in near-real-time after earthquakes. A M6.5 occurring in a populated region may be felt by 10–50 million people at intensities ranging from barely felt (MMI II–III) to damaging (MMI VII+). The 2011 Tohoku M9.1 earthquake was felt across the entire Japanese archipelago (population ~127 million) and recorded instrumentally in North America, Europe, and Australia. USGS 'Did You Feel It?' (DYFI) reports provide crowd-sourced intensity data; major California earthquakes regularly generate hundreds of thousands of DYFI reports within 30 minutes, demonstrating the extraordinary geographic reach of felt shaking.

Can animals detect earthquakes before humans feel them?

Anecdotal reports of unusual animal behavior before earthquakes have a multi-century history and have been studied scientifically with mixed results. Some well-documented mechanisms could theoretically allow animal pre-earthquake detection: radon gas emanating from stressed rock before rupture, electromagnetic signals from piezoelectric effects in quartz-bearing rock, and changes in well water levels. The 2011 Peru study (Nature, Bleier et al.) documented behavioral changes in dogs correlating with seismic events in the preceding days. However, controlled prospective studies—designed to distinguish genuine precursory behavior from random behavioral variation—have not produced reliable results. The scientific consensus is that while physical precursors sometimes precede earthquakes, no animal behavior pattern has demonstrated predictive skill sufficient for operational early warning.

What is seismic early warning and how does it differ from prediction?

Earthquake early warning (EEW) and earthquake prediction are fundamentally different concepts. EEW detects the initial, less-damaging P-waves from a seismic event seconds after rupture begins and transmits automated alerts before the slower, more-destructive S-waves and surface waves arrive at distant locations. ShakeAlert, deployed by USGS across the western US, provides 2–90 seconds of warning depending on distance from the epicenter—enough time to stop trains, open fire station doors, or adopt protective posture. Japan's J-ALERT system provides up to 60–90 seconds warning for distant events. Earthquake prediction, by contrast, attempts to specify the time, location, and magnitude of a future earthquake before it occurs—a goal that remains scientifically unachieved for deterministic forecasts on timescales of days to years.