구텐베르크-리히터 법칙: 빈도-규모 관계

The Gutenberg-Richter Formula: log10(N) = a − bM

The Gutenberg-Richter LawA statistical law describing the relationship between earthquake frequency and magnitude: for each unit increase in magnitude, earthquakes become about 10 times less frequent. relationship is one of the most fundamental empirical laws in seismology. Formulated by Beno Gutenberg and Charles Richter in 1944 from a study of earthquake catalogs in southern California, it states that the number of earthquakes N with MagnitudeA single number that quantifies the total energy released by an earthquake. Each whole number increase represents roughly 31.6 times more energy released. greater than or equal to M follows the logarithmic relationship: log₁₀(N) = a − bM, where a and b are constants. The parameter a describes the overall level of seismicity in a region — larger a means more earthquakes of all magnitudes. The parameter b — the b-valueThe slope of the Gutenberg-Richter frequency-magnitude relationship. A b-value near 1.0 is typical; higher values indicate more small earthquakes relative to large ones. Changes may signal stress changes. — describes the relative proportion of small earthquakes to large ones. A b-value of 1.0 (the typical global average) means that for every increase of one magnitude unit, there are ten times fewer earthquakes. For every Mw 7.0 earthquake, there should be roughly ten Mw 6.0 earthquakes, one hundred Mw 5.0 earthquakes, and one thousand Mw 4.0 earthquakes. This elegant regularity holds over an extraordinary range — from magnitude 2 microearthquakes to the largest recorded megathrust events — and has profound implications for Seismic Risk AssessmentThe process of evaluating earthquake hazard, building vulnerability, and potential losses for a specific area or structure. Combines hazard maps, building inventory, and damage models. and Probabilistic Seismic Hazard Analysis (PSHA)A method for quantifying earthquake hazard that considers all possible earthquake sources, magnitudes, and ground motion levels, expressing results as probability of exceeding specific shaking levels..

Physical Interpretation of the Relationship

The Gutenberg-Richter law reflects the self-similar, fractal geometry of fault networks. Small earthquakes rupture small fault patches; large earthquakes rupture larger ones. The distribution of fault patch sizes in a fault network follows a power-law distribution, which directly produces the Gutenberg-Richter log-linear relationship between frequency and MagnitudeA single number that quantifies the total energy released by an earthquake. Each whole number increase represents roughly 31.6 times more energy released.. The b-value encodes information about the stress state of the crust and the geometry of the fault network: high stress tends to produce lower b-values (relatively more large earthquakes), while low stress or thermally weakened rock produces higher b-values.

What the b-value Reveals

The b-valueThe slope of the Gutenberg-Richter frequency-magnitude relationship. A b-value near 1.0 is typical; higher values indicate more small earthquakes relative to large ones. Changes may signal stress changes. is not merely a fitting parameter — it carries physical information about the state of the seismogenic crust. In the global earthquake catalog, b is remarkably close to 1.0, but regional and local variations are meaningful and diagnostic. Low b-values (below 0.9) are associated with regions of high tectonic stress, cold, rigid continental crust, and the locked zones of major Subduction ZoneA region where one tectonic plate dives beneath another into the mantle. Subduction zones produce the world's largest earthquakes (M8.5+) and are associated with deep ocean trenches and volcanic arcs.s. They indicate that the local fault network tends to produce relatively more large earthquakes relative to small ones. High b-values (above 1.2) are found in geothermal areas, volcanic regions, fluid-saturated fault zones, and areas where pore fluid pressure is elevated. The connection between elevated pore pressure and high b-values is particularly important for understanding Induced SeismicityEarthquakes triggered by human activities such as hydraulic fracturing (fracking), wastewater injection, mining, or reservoir impoundment. Most are small (M<4) but some have exceeded M5.5.: wastewater injection raises pore fluid pressure in the crust, reducing effective stress and typically increasing the b-value of the induced seismicity, which can serve as a diagnostic indicator.

Temporal Variations in b-value

The b-valueThe slope of the Gutenberg-Richter frequency-magnitude relationship. A b-value near 1.0 is typical; higher values indicate more small earthquakes relative to large ones. Changes may signal stress changes. can also vary through time. During earthquake swarmsA sequence of earthquakes occurring in a localized area over days to months with no clearly dominant mainshock. Often associated with volcanic activity or fluid injection., the b-value is often notably higher than during background seismicity. In the post-mainshock period, the b-valueThe slope of the Gutenberg-Richter frequency-magnitude relationship. A b-value near 1.0 is typical; higher values indicate more small earthquakes relative to large ones. Changes may signal stress changes. of the AftershockA smaller earthquake that follows the mainshock in the same fault region. Aftershock sequences can last weeks to years, with the largest aftershock typically 1.0-1.2 magnitudes below the mainshock. sequence is typically slightly different from the background level. Some researchers have proposed that systematic decreases in the b-valueThe slope of the Gutenberg-Richter frequency-magnitude relationship. A b-value near 1.0 is typical; higher values indicate more small earthquakes relative to large ones. Changes may signal stress changes. measured on a fault over months to years might signal increasing stress and an elevated probability of a future large earthquake — essentially a potential precursor. While this idea is theoretically motivated and has some empirical support, its practical utility for earthquake forecastingPrediction claims to specify exact time, place, and magnitude of a future earthquake — currently impossible. Forecasting provides probabilistic estimates of earthquake likelihood over time periods. remains controversial because b-value estimation requires large earthquake catalogs and is subject to significant statistical uncertainty.

Regional Variations in the b-value

The geographic distribution of b-values across global and regional earthquake catalogs reveals systematic patterns tied to tectonic environment. Subduction zones typically show b-values near 0.8–1.0, with the locked megathrust interface often exhibiting the lowest values, reflecting the high compressive stress in those environments. Mid-ocean ridges and back-arc basins tend toward higher b-values, reflecting the extensional, thermally elevated regime. Volcanic arcs have notably high b-values, especially in the edifice zones where magma heating weakens the rock. In the continental United States, the Pacific Coast seismic zones have b-values near 1.0, while the stable continental interior shows somewhat higher values. The spatial variation in b-valueThe slope of the Gutenberg-Richter frequency-magnitude relationship. A b-value near 1.0 is typical; higher values indicate more small earthquakes relative to large ones. Changes may signal stress changes. across a seismogenic region can be mapped using modern catalog data and statistical methods, producing a b-value anomaly map that highlights areas of elevated stress — a tool used in seismic hazard assessment.

b-value in Induced Seismicity Monitoring

Monitoring the b-valueThe slope of the Gutenberg-Richter frequency-magnitude relationship. A b-value near 1.0 is typical; higher values indicate more small earthquakes relative to large ones. Changes may signal stress changes. of Induced SeismicityEarthquakes triggered by human activities such as hydraulic fracturing (fracking), wastewater injection, mining, or reservoir impoundment. Most are small (M<4) but some have exceeded M5.5. associated with industrial operations — wastewater injection, hydraulic fracturing, geothermal energy production — has become a practical tool for traffic light protocols (TLPs) that govern whether operations should continue, be modified, or cease. A sudden decrease in the b-valueThe slope of the Gutenberg-Richter frequency-magnitude relationship. A b-value near 1.0 is typical; higher values indicate more small earthquakes relative to large ones. Changes may signal stress changes. of a monitored induced seismicity sequence may indicate that stress is building toward a larger event, warranting precautionary action. Conversely, a high b-valueThe slope of the Gutenberg-Richter frequency-magnitude relationship. A b-value near 1.0 is typical; higher values indicate more small earthquakes relative to large ones. Changes may signal stress changes. suggests that the induced earthquakes are occurring in a low-stress, high-pore-pressure environment where very large events are less likely. These monitoring applications of the Gutenberg-Richter LawA statistical law describing the relationship between earthquake frequency and magnitude: for each unit increase in magnitude, earthquakes become about 10 times less frequent. law are actively used by operators and regulators in Oklahoma, Kansas, Ohio, and other US states where injection-induced seismicity has become a significant public concern.

Applications in Seismic Hazard Assessment

The Gutenberg-Richter relationship is a cornerstone of Probabilistic Seismic Hazard Analysis (PSHA)A method for quantifying earthquake hazard that considers all possible earthquake sources, magnitudes, and ground motion levels, expressing results as probability of exceeding specific shaking levels.. In Probabilistic Seismic Hazard Analysis (PSHA)A method for quantifying earthquake hazard that considers all possible earthquake sources, magnitudes, and ground motion levels, expressing results as probability of exceeding specific shaking levels., each seismic source — whether a mapped Fault (Geology)A fracture in rock along which movement has occurred. Faults range from millimeters to thousands of kilometers long. Major faults that produce earthquakes are called active faults. with its own characteristic magnitude distribution, or a distributed area source capturing background seismicity — is assigned a magnitude-frequency distribution. For distributed seismicity sources, the Gutenberg-Richter distribution with locally calibrated a and b-valueThe slope of the Gutenberg-Richter frequency-magnitude relationship. A b-value near 1.0 is typical; higher values indicate more small earthquakes relative to large ones. Changes may signal stress changes. is the standard model. The rate of earthquake occurrence at any given magnitude is read directly from this distribution, and the resulting rate is fed into the hazard integral that combines source rates, ground motion prediction equations, and distance attenuation to produce the final hazard curve at a site. The accuracy of Probabilistic Seismic Hazard Analysis (PSHA)A method for quantifying earthquake hazard that considers all possible earthquake sources, magnitudes, and ground motion levels, expressing results as probability of exceeding specific shaking levels. depends critically on the accuracy of the Gutenberg-Richter parameters: errors in b of even ±0.1 can significantly affect hazard estimates, especially at the large-magnitude tail of the distribution where the rarest and most damaging events are located.

Limits and Exceptions to the Law

The Gutenberg-Richter relationship is not universally applicable in its simplest form. At low magnitudes, the catalog becomes incomplete because small earthquakes are below the detection threshold of any Seismic NetworkA coordinated group of seismograph stations that continuously monitor earthquake activity. The Global Seismographic Network (GSN) includes 150+ stations providing worldwide coverage. — a minimum magnitude of completeness Mc must be determined before fitting the Gutenberg-Richter parameters. At large magnitudes, the distribution must be truncated at some maximum magnitude Mmax, representing the largest possible earthquake on a given source; without this upper bound, the integral of the Gutenberg-Richter distribution diverges. Some fault systems appear to produce characteristic earthquakes — repeated ruptures of nearly the same magnitude and length — rather than the smooth Gutenberg-Richter distribution of sizes. The evidence for characteristic earthquake behavior versus Gutenberg-Richter behavior is debated, with implications for how hazard on major faults should be calculated. The 2011 Tohoku earthquake, which significantly exceeded the Mw 8.4 maximum assumed in Japanese national hazard models, is a sobering reminder that apparent catalog limits may reflect the finite length of the historical record rather than true physical upper bounds on earthquake MagnitudeA single number that quantifies the total energy released by an earthquake. Each whole number increase represents roughly 31.6 times more energy released..