Fault Trenching Simulator: Reading Earthquake History in Stratigraphy

simulator intermediate ~10 min
Loading simulation...
D_total = 8 m — 4 events at 2 m each

Four earthquake events with 2 m displacement each produce 8 m of cumulative offset visible in the trench wall. At 1 mm/yr sedimentation, each event horizon is separated by distinct sediment packages datable by radiocarbon.

Formula

D_total = D × N (cumulative displacement)
RI_avg = age_span / (N - 1) (average recurrence interval)
slip_rate = D_total / age_span (long-term slip rate)

Excavating the Past

A paleoseismic trench is an archaeological dig for earthquakes. By cutting a 3–5 m deep excavation perpendicular to an active fault, geologists expose a vertical cross-section through thousands of years of earthquake history. Each rupture event displaces sediment layers, creates a surface scarp that erodes into a distinctive wedge-shaped deposit, and is eventually buried by ongoing sedimentation. Reading this stratigraphic record requires the skills of both a structural geologist and a sedimentologist.

Event Identification

Recognizing individual earthquakes in a trench wall is detective work. Key indicators include fault strands that terminate upward at a specific horizon (the event horizon), colluvial wedges of scarp-derived sediment burying the pre-event surface, fissure fills where coseismic cracks opened and filled with surface material, and abrupt changes in layer thickness across the fault. Each indicator alone is ambiguous; together they build a convincing case for a specific event.

Dating the Record

Radiocarbon dating of detrital charcoal, seeds, or organic-rich sediment above and below event horizons brackets earthquake ages. With careful stratigraphic ordering and Bayesian statistical models (like OxCal), paleoseismologists can constrain event dates to within decades, even for earthquakes thousands of years old. The sedimentation rate determines how much material accumulates between events and thus how resolvable the record is.

From Trench to Hazard

Trench data feed directly into seismic hazard analysis. The number of events, their ages, and displacement per event constrain recurrence intervals, slip rates, and characteristic earthquake magnitudes for a fault. This simulation lets you build a virtual trench — adjusting displacement, sedimentation, and erosion to understand how the geological record captures (or fails to capture) the earthquake history.

FAQ

What is fault trenching in paleoseismology?

Fault trenching involves excavating a trench perpendicular to an active fault to expose the vertical cross-section of displaced sediment layers. Each past earthquake that ruptured the surface offsets layers and creates distinctive features: fault scarps, colluvial wedges, fissure fills, and disrupted horizons. By dating these features, scientists reconstruct the earthquake history.

How deep are paleoseismic trenches?

Typical paleoseismic trenches are 3–5 m deep, 1–2 m wide, and 20–50 m long. Depth depends on the cumulative displacement to be investigated and the burial depth of the oldest event horizons. Some exceptionally deep trenches reach 8–10 m using benched excavation for safety.

How are earthquake events identified in trench walls?

Events are identified by: upward termination of fault strands at a specific horizon, colluvial wedges (scarp-derived debris) burying the pre-event surface, fissure fills crossing older deposits, warped or tilted layers, and liquefaction features. Each event is bounded by an 'event horizon' representing the ground surface at the time of the earthquake.

How are paleoseismic events dated?

Radiocarbon dating of organic material (charcoal, seeds, shells) above and below event horizons brackets the earthquake age. Optically stimulated luminescence (OSL) dates sediment burial time. Dendrochronology and tephrochronology provide additional constraints. Bayesian statistical models combine multiple dates to refine event chronologies.

Sources

Embed

<iframe src="https://homo-deus.com/lab/paleoseismology/fault-trenching/embed" width="100%" height="400" frameborder="0"></iframe>
View source on GitHub