A finite-fault model represents an earthquake as slip distributed across a fault surface rather than as one point at the hypocentre. Seismologists divide a possible fault plane into smaller patches and use seismic waveforms, ground displacement and other observations to estimate how much each patch slipped and, in a kinematic model, when that slip occurred. The result can reveal rupture length, high-slip areas, propagation and moment release. ShakeMap uses earthquake observations and ground-motion models to estimate shaking across a region. When a credible finite rupture becomes available, its geometry improves distance calculations and the spatial pattern compared with a point source, especially for large earthquakes. Both products are models. Rapid versions may change as stations report, the causative fault plane becomes clearer and more complete inversions replace early assumptions.
Key takeaways
- Large earthquakes rupture an area, not a mathematical point.
- Finite-fault inversion estimates slip on many fault patches.
- Static models estimate slip; kinematic models also estimate timing.
- ShakeMap combines observations, rupture geometry and ground-motion models.
- Early products may use a point source before finite geometry is available.
- Model updates reflect new evidence and should retain version history.
Why the point-source picture is limited
An epicentre is useful for locating rupture initiation, but a major fault can extend tens or hundreds of kilometres. Communities near different parts of that fault have different distances and wave radiation. Treating the event as a point can distort shaking estimates, exposure and aftershock context. Finite geometry becomes increasingly important as rupture size grows.
The fault plane and its patches
Analysts choose one or more fault surfaces with strike, dip and depth, then divide them into subfaults. Each patch can have a slip amplitude and direction. A kinematic inversion adds rupture initiation and rise time. The grid is a computational representation; patch boundaries do not imply that nature slipped in neat rectangles.
The two possible focal-mechanism planes
A moment-tensor solution commonly produces two geometrically possible planes: the physical fault plane and an auxiliary plane. Tectonic setting, aftershocks, mapped rupture and geodetic motion help identify the causative plane. When evidence is weak, analysts test both. Choosing too early can build an elegant slip pattern on the wrong geometry.
What data constrain the model
Teleseismic body and surface waves record rupture from many directions. Nearby strong-motion stations capture intense shaking. GNSS measures permanent displacement, while InSAR maps line-of-sight surface deformation. Tsunami and field observations may add constraints. Each dataset samples different aspects and has its own noise, coverage and modelling assumptions.
Inversion rather than direct imaging
The slip pattern is not photographed underground. Analysts calculate how a trial distribution would affect observations, compare predictions with data and adjust model parameters. Many distributions can fit imperfect data. Regularisation and physical constraints prevent implausibly rough solutions, but they also influence the final appearance. The model should be read with its assumptions and resolution tests.
Static and kinematic finite-fault models
A static model estimates the final slip distribution. A kinematic model also describes how rupture propagated and how long patches slipped. Kinematic results support source-time functions and directivity analysis but require observations that contain time information. A detailed animation does not mean every frame is uniquely resolved by the data.
Seismic moment and magnitude
Seismic moment depends on fault area, average slip and rock rigidity. The moment contributed by every patch sums to the event total, which is converted to moment magnitude. Two earthquakes with similar magnitude can have different slip concentration and rupture geometry, producing different patterns of shaking and deformation.
Rupture speed and rise time
Rupture speed describes how the failure front moves across the fault; rise time is how long a patch continues slipping. These affect waveform shape and directivity. USGS workflows test a range of velocities and may allow variation around a preferred value. Sparse station coverage can leave trade-offs between timing, geometry and slip.
What ShakeMap estimates
ShakeMap produces spatial estimates of ground-motion measures and intensity soon after an earthquake. It blends recorded values with predictions that account for magnitude, distance, site conditions and regional models. It is not a map of fault slip, damage or aftershocks, although those products are related and can support response decisions.
How rupture geometry enters ShakeMap
Distance to a finite rupture is more meaningful than distance to the epicentre for a large event. USGS ShakeMap can use a rupture polygon from a finite-fault solution. When none is available, it uses a point source and may simulate plausible fault realisations. Adding a finite model affects distance metrics, data handling and the predicted spatial pattern.
Stations still anchor the map
A rupture model does not replace measured shaking. Instrumented records and community intensity reports constrain the map where available. In regions with sparse stations, the prediction model and site map carry more weight. Users should distinguish areas supported by observations from areas filled mainly by modelling.
Site effects and basin response
Soft sediment, basin geometry, topography and shallow geology can amplify or prolong motion independently of the source slip. ShakeMap incorporates available site information, but local three-dimensional effects may exceed map resolution. A realistic finite fault improves source distance without solving every uncertainty in the propagation path or local ground.
Why rapid models change
The first minutes may provide only hypocentre, magnitude and focal mechanism. More stations, revised locations, GNSS, InSAR and field mapping arrive later. Analysts may change fault plane, segmentation or rupture extent. An update is not evidence that earlier science was dishonest; rapid products trade completeness for speed and should display timestamps and versions.
Resolution and uncertainty
Slip images often use vivid colour scales that look exact. In reality, neighbouring patches may trade slip, offshore sections may be poorly observed and assumed Earth structure can bias results. Checkerboard or synthetic tests, alternative geometries and data-fit plots reveal what is resolved. Response users need robust features, not every coloured pixel.
From maps to response
Emergency managers use shaking estimates to prioritise reconnaissance, estimate exposed population and identify transport or utility corridors needing inspection. PAGER adds loss estimation. Decisions should be revised as products update and combined with direct reports. A model can focus attention quickly, but it cannot certify that an unreported community is unharmed.
A classroom inversion exercise
Students can divide a paper fault into patches, assign hidden slip and create simplified “station” sums. Classmates try to reconstruct the pattern from limited observations. Adding more stations or permanent-displacement data narrows possibilities, while noise produces several acceptable answers. The exercise explains why inversion is powerful but not identical to direct observation.
Data weighting and model trade-offs
An inversion combines datasets with different units, noise and numbers of observations. If thousands of waveform samples overwhelm a smaller GNSS dataset numerically, the model may fit waves while missing permanent displacement. Analysts choose weights and filters to balance information. Those choices are part of the model and should be tested. A good overall fit can still conceal poor performance for one dataset, so separate residual plots are more informative than a single summary score.
Fault segmentation and complex rupture
Real earthquakes can bend, branch or jump between mapped faults. A single rectangle may be adequate for rapid analysis but fail to represent a multi-segment rupture. Adding segments increases realism and the number of parameters, which can exceed what data resolve. Field rupture, aftershocks and geodesy help justify complexity. The simplest model that explains robust observations is often more useful for response than a detailed pattern dominated by assumptions.
Connecting finite faults to tsunamis
For undersea earthquakes, seafloor displacement from the rupture can initialise tsunami models. Vertical motion, fault depth and geometry matter more than magnitude alone. A rapid finite model can improve warning scenarios, but tsunami centres also use sea-level observations to confirm waves. Landslides and complex bathymetry introduce additional uncertainty. The earthquake inversion is therefore an important input, not a complete forecast of coastal height or arrival.
Communicating updates during a crisis
Maps should display issue time, model version and a concise explanation of what changed. Emergency users need to know whether an update moved the rupture, added stations or altered site corrections. Archiving earlier products supports after-action review while avoiding simultaneous circulation of unlabeled versions. Public explanations should emphasise that refinement is expected: speed is valuable in the first minutes, and accuracy improves as a larger evidence set becomes available.
Aftershocks and stress change
Finite slip models can be used to calculate changes in stress on nearby faults and compare them with aftershock patterns. These calculations depend on assumed fault geometry, rock properties and the chosen slip distribution. They identify regions where failure may have been encouraged or discouraged, not the time of the next earthquake. Aftershocks also refine the rupture outline but do not exactly trace every slipped patch. Forecast communication should avoid converting a probabilistic stress calculation into a deterministic prediction.
A checklist for map users
Before acting on a map, check the product type, issue time, magnitude and rupture version. Identify which areas contain instruments and where estimates rely on models. Read the legend carefully: intensity, acceleration, velocity, slip and probability are different quantities. Note site-condition resolution and whether uncertainty is displayed. For operational decisions, combine the map with direct reports and update alerts. For research, retain the exact version because a later download may contain revised geometry and data.
Questions to ask before accepting the conclusion
For any claim about finite-fault model, identify what was measured directly, which steps relied on models or judgement, and whether the sampled places and times represent the wider question. Check the units, comparison group, missing observations and uncertainty range. Look for an independent method with different weaknesses and ask whether it supports the same central result. Finally, separate a useful screening signal from evidence strong enough for an expensive or irreversible decision. This checklist does not demand perfect certainty; it ensures that the confidence of the conclusion matches the quality and scope of the evidence.
Related Earth Site reading
Useful background is available in rupture directivity, earthquake intensity maps, locating an earthquake and what causes earthquakes. These public Earth Site pages were checked as internal destinations; the independent sources listed in the companion notes support the research in this article.
Conclusion
Finite-fault models reveal an earthquake as an extended rupture with uneven slip and evolving moment release. ShakeMap uses that geometry, together with observations and ground-motion models, to estimate where shaking was strongest. The products answer different questions and improve one another, yet both remain conditional on data and assumptions. Rapid versions should guide urgent attention while later versions refine the scientific record. Reading their timestamps, data coverage and uncertainty prevents a compelling map from being mistaken for a complete measurement of every place and every part of the fault.
Frequently asked questions
Is a finite-fault model a photograph of the fault?
No. It is an inversion that estimates slip from observations and physical calculations.
Does ShakeMap show where the fault slipped most?
Not directly. It maps estimated ground shaking; finite-fault slip is a separate input and product.
Why does an earthquake map change after release?
More data and improved rupture geometry become available, so estimates are recalculated.
Is the epicentre the centre of the whole rupture?
Not necessarily. It is the surface point above rupture initiation.
Are finite-fault models produced for small earthquakes?
They are most useful and resolvable for larger events with sufficient observations.
