The Physics of Seismic Waves and S−P Intervals
When an earthquake occurs, it releases energy in the form of seismic waves that travel through the Earth. Among these, body waves provide the primary data used to locate the event's origin. Primary waves (P-waves) are compressional waves that travel fastest through geological materials. Secondary waves (S-waves) are shear waves that travel slower than P-waves. Because P-waves travel faster than S-waves, the time interval between their arrivals at a recording station—known as the S−P interval—grows larger the farther the waves travel from the source.
Under a constant-speed assumption, this arrival-time difference satisfies the relationship:
Δt = D(1/Vs − 1/Vp)
By rearranging this formula, we can calculate the distance (D) from the recording station to the earthquake source based on the time delay (Δt) between the wave arrivals:
D = Δt ÷ (1/Vs − 1/Vp)
This relationship forms the foundation of seismic triangulation. By measuring the S−P interval at multiple stations and applying known wave speeds, seismologists can determine how far the earthquake occurred from each station.
The Geometry of Triangulation
Locating an epicenter requires observations from multiple geographic points. A single recording station provides only an S−P interval, which translates to a calculated radius. This radius defines a circle on the Earth's surface where the earthquake could have occurred, but it provides no direction.
When a second station is added, its distance circle will intersect the first station's circle at two distinct points. To resolve this ambiguity and identify the unique epicenter, a third station is required.
While three stations are the standard educational construction for locating an epicenter, adding more stations (up to eight) allows for a more robust calculation. When real-world station data are used, the circles rarely intersect at a single, perfect point due to timing uncertainties and geological variations. Using four or more stations makes these mismatches visible, allowing for a mathematical optimization of the estimated epicenter.
Least-Squares Optimization on a Spherical Earth
To reconcile imperfect observations from multiple stations, the Earthquake Epicenter Triangulation Calculator uses a least-squares treatment. Instead of assuming all distance circles intersect perfectly, the tool minimizes the sum of the squared differences between the calculated S−P distances and the actual spherical surface distances from the estimated epicenter to each station.
The optimization process relies on the following mathematical formulations:
- Distance approximation: D = Δt ÷ (1/Vs − 1/Vp)
- Surface distance: d = R × haversine central angle, with R = 6,371 km
- Epicenter fit: minimize Σ(dᵢ − Dᵢ)² across all stations
The Earth is modeled as a sphere with a constant radius (R) of 6,371 km. The tool calculates a Distance factor to convert S−P intervals into distances:
Distance factor = Vp × Vs ÷ (Vp − Vs)
For each station, the individual distance derivation is computed as:
Station S−P interval × Distance factor = Calculated radius
The algorithm searches for the coordinates that yield the Best spherical fit by minimizing the Root Mean Square (RMS) distance residual.
Analyzing Residuals and Station Geometry
Once the least-squares fit is complete, the tool evaluates the quality of the solution using residuals and geometric strength. A residual is calculated as the fitted spherical distance minus the distance inferred from the S−P interval.
- Positive Residual: The fitted epicenter is farther from the station than its calculated S−P radius.
- Negative Residual: The fitted epicenter is closer to the station than its calculated S−P radius.
The tool reports the RMS distance residual and the Largest distance residual to indicate how well the observations align. The overall fit is categorized under Distance fit quality as either a close distance fit, mixed distance fit, or poor distance fit.
The spatial distribution of the stations relative to the epicenter—known as station geometry—strongly influences the stability of the solution. Stations that are well-distributed around the source constrain both latitude and longitude effectively. If stations are clustered together or arranged in a straight line, the geometry is weak, making the calculated epicenter highly sensitive to small timing errors. The tool rates this configuration under Station geometry quality as strong station geometry, moderate station geometry, or weak station geometry.
Error Propagation and Model Simplifications
The calculator reports an Approximate position 1σ and Distance 1σ per station based on the user-entered Timing uncertainty (1σ). This timing uncertainty propagates through the station geometry to define an uncertainty boundary along a major horizontal axis.
It is important to recognize that this tool is designed as an interactive teaching model and uses several simplifying assumptions:
- Spherical Earth: It assumes a uniform spherical surface with a constant radius of 6,371 km.
- Constant Wave Speeds: It assumes P-wave and S-wave velocities are constant throughout the medium, omitting layered or three-dimensional velocity structures.
- Shallow-Source Approximation: It assumes a shallow-source earthquake, omitting the focal depth of the event.
- No Systematic Errors: The calculated uncertainty propagates only the timing value entered by the user. It does not account for velocity-model errors, phase misidentification, clock bias, or curved ray paths.
For professional seismic-catalog locations, seismologists must use complex travel-time models that account for depth, distance, and the specific layered structures of the Earth's crust and mantle.
Using the Calculator
To use the calculator, enter the coordinates and S−P intervals for three to eight stations. You must also specify the propagation model parameters, including the P-wave speed, S-wave speed, and timing uncertainty.
Input Constraints and Validation
- Station Limits: You must use between three and eight stations. If you attempt to calculate with fewer than 3 or more than 8, the tool displays:
Use between three and eight stations. - Latitude: Must be from −90° to 90°. Out-of-bounds entries return:
Station station: latitude must be from −90° to 90°. - Longitude: Must be from −180° to 180°. Out-of-bounds entries return:
Station station: longitude must be from −180° to 180°. - S−P interval: Must be greater than 0 and no more than 2,000 seconds. Out-of-bounds entries return:
Station station: enter an S−P interval greater than 0 and no more than 2,000 seconds. - Wave Speeds: Both P-wave and S-wave speeds must be greater than 0 and no more than 20 km/s. Out-of-bounds entries return:
field must be greater than 0 and no more than 20 km/s.Additionally, the P-wave speed must be greater than the S-wave speed, or the tool displays:P-wave speed must be greater than S-wave speed. - Timing Uncertainty: Must be greater than 0 and no more than 60 seconds. Out-of-bounds entries return:
Timing uncertainty must be greater than 0 and no more than 60 seconds. - Duplicate Stations: If two stations share identical coordinates, the tool displays:
Stations station and other have the same coordinates. Use distinct station locations. - Unstable Geometry: If the stations are arranged in a straight line or tight cluster, the tool displays:
These stations do not constrain a stable two-dimensional location. Add a station away from the current line or cluster. - Non-Numeric Inputs: If a non-numeric value is entered, the tool displays:
where: “token” is not a number.
You can click Load Tokyo example to populate the tool with sample data, or click Clear to reset all inputs.
Privacy and Data Processing
All station coordinates, S−P times, and calculation results are processed locally within your web browser. No data is uploaded to external servers.