古地震学(paleoseismology): 岩石から地震の歴史を読む

What Is Paleoseismology?

古地震学断層トレンチ、隆起した段丘、津波堆積物といった地質学的証拠を通じて、先史時代の地震を研究する学問。地震の記録を数千年前まで遡らせる。 is the branch of earthquake science that studies prehistoric earthquakes using geological evidence preserved in the landscape and in the sedimentary record. Because instrumental and historical earthquake records extend back at most a few hundred years — a tiny fraction of the recurrence intervals of large earthquakes on many faults — paleoseismology is essential for understanding the long-term behavior of fault systems. By identifying, dating, and measuring the physical evidence of past earthquakes preserved in sediments, bedrock, and landforms, paleoseismologists can extend the earthquake record thousands to tens of thousands of years into the past, revealing patterns of rupture that would be invisible in the short instrumental record. This extended record is fundamental to probabilistic seismic hazard analysis (PSHA) and to 地震空白域隣接する区間と比較して、長期間にわたり地震が発生していない活断層の区間。地震空白域は、将来の地震発生確率が高いことを示唆する場合がある。 identification.

The Value of Long Records

The fundamental challenge in earthquake science is that the largest, most dangerous earthquakes occur infrequently. A 断層岩盤に沿って動きが生じた破断面。断層は数ミリメートルから数千キロメートルまで様々な長さがある。地震を引き起こす主要な断層は活断層と呼ばれる。 that generates Mw 8.0 earthquakes every 500 years will have produced only one or two events in the entire period covered by modern seismograph networks (roughly 1900 to present). To estimate how often such an earthquake will occur in the future — the foundation of 地震リスク評価特定の地域や構造物について、地震ハザード・建物の脆弱性・想定される損失を評価する過程。ハザードマップ、建物台帳、被害モデルを組み合わせて行われる。 — we need many past events, not just one or two. 古地震学断層トレンチ、隆起した段丘、津波堆積物といった地質学的証拠を通じて、先史時代の地震を研究する学問。地震の記録を数千年前まで遡らせる。 provides this long record, turning the geology of fault zones into a detailed history of earthquake activity stretching across many seismic cycles.

Fault Trench Studies: Reading the Layers

The most widely used technique in 古地震学断層トレンチ、隆起した段丘、津波堆積物といった地質学的証拠を通じて、先史時代の地震を研究する学問。地震の記録を数千年前まで遡らせる。 is the excavation of trenches across active 断層線地表に現れた断層の痕跡で、線状または破砕された岩石の帯として視認できる。地質学者は活断層線を地図化し、周辺地域の地震リスクを評価する。s. A trench typically cuts perpendicular to the fault trace, 1–3 meters wide and 2–5 meters deep, extending far enough on each side to expose undisturbed stratigraphy away from the fault zone. The walls of the trench are carefully logged — every sediment layer, soil horizon, and fault strand is mapped in detail. Earthquakes leave characteristic signatures in the stratigraphy: colluvial wedges (masses of sediment shed from a 断層崖地震時の断層に沿った垂直方向の変位によって形成される崖や急斜面。断層崖は高さ数メートルに達することがあり、過去の地震活動を示す目に見える証拠となる。 after it is uplifted by an earthquake), buried soil horizons truncated by fault displacement, fissures filled with dike-like injections of sand or mud from below, and angular unconformities where shaking-liquefied sediments were injected upward. Each such event horizon, if it can be confidently identified and dated, represents a past earthquake, providing a time series of rupture events going back as far as the preserved stratigraphy extends.

The Pallett Creek Record

The most extensively studied paleoseismic site in the world is Pallett Creek on the Mojave segment of the San Andreas 断層線地表に現れた断層の痕跡で、線状または破砕された岩石の帯として視認できる。地質学者は活断層線を地図化し、周辺地域の地震リスクを評価する。 in southern California. Kerry Sieh's landmark 1978 study documented 12 large earthquakes at this site over the past 2,000 years, including the great 1857 Fort Tejon earthquake. Subsequent work extended the record further and refined the dates. The Pallett Creek record demonstrates that the southern San Andreas does not behave with perfectly periodic repetition — the intervals between events range from roughly 50 to 330 years — but on average produces a major rupture every 130–140 years. The site last ruptured in 1857; this long quiescence, combined with the 固着断層摩擦によって動きが妨げられ、応力が蓄積している断層区間。固着断層がついに破壊すると、大地震を引き起こすことがある。 character of the segment, indicates significant accumulated strain.

Dating Past Earthquakes with Radiocarbon

Radiocarbon (¹⁴C) dating is the workhorse technique for determining the ages of organic materials preserved in fault trench stratigraphy. Carbon-14 is produced in the atmosphere by cosmic ray bombardment of nitrogen-14 and incorporated into living organisms during their lifetimes. When an organism dies, it no longer takes in new carbon, and the ¹⁴C in its tissues decays with a half-life of 5,730 years. By measuring the ratio of ¹⁴C to stable ¹²C in charcoal, plant remains, or shell fragments from sediment layers above and below an earthquake horizon, the ages of those layers can be determined, bracketing the time of the earthquake. Radiocarbon dating is most useful for earthquakes younger than about 50,000 years and achieves precision of roughly ±50–200 years with modern accelerator mass spectrometry (AMS) techniques. Other dating methods used in 古地震学断層トレンチ、隆起した段丘、津波堆積物といった地質学的証拠を通じて、先史時代の地震を研究する学問。地震の記録を数千年前まで遡らせる。 include optically stimulated luminescence (OSL), cosmogenic nuclide exposure dating, and U-series dating of carbonate cements.

Uplifted Terraces and Tsunami Deposits

Not all paleoseismic evidence comes from fault trenches. Coastal geomorphology preserves a rich record of past earthquakes and tsunamis. Uplifted marine terraces — flat benches carved by wave action at sea level during past time periods and now elevated above the ocean — record co-seismic uplift during past subduction earthquakes. The coast of the Huon Peninsula in Papua New Guinea preserves a staircase of uplifted terraces recording repeated earthquakes over hundreds of thousands of years. In Cascadia, ghost forests of drowned trees — standing snags preserved in coastal marshes — record sudden coseismic subsidence during past megathrust earthquakes; radiocarbon dating of these trees has precisely dated the last Cascadia megathrust to January 26, 1700, based on Japanese tsunami records. 津波海底地震時の海底の急激な変位によって発生する一連の海の波。津波はジェット機並みの速度（時速700km以上）で海洋全域を伝わることがある。 deposits — anomalous layers of sand and marine microfossils deposited inland by past tsunamis — independently confirm the occurrence and size of past megathrust earthquakes, complementing the fault trench evidence.

Extending the Earthquake Record: Why It Matters for Recurrence Interval

The practical importance of 古地震学断層トレンチ、隆起した段丘、津波堆積物といった地質学的証拠を通じて、先史時代の地震を研究する学問。地震の記録を数千年前まで遡らせる。 lies in its contribution to earthquake hazard assessment. The 地震再来間隔特定の断層における大地震と大地震の間の平均時間。古地震学や歴史記録から推定される。カスケード沈み込み帯の再来間隔は約500年である。 of a fault — the average time between large earthquakes — cannot be reliably estimated from a record of only one or two events. With a paleoseismic record covering 10 or more events, the mean recurrence interval can be estimated with reasonable confidence, and the variability in that interval can be characterized. This variability is important: the "time-predictable" and "slip-predictable" models assume different relationships between recurrence and slip, and the choice of model affects hazard calculations significantly. 地震空白域隣接する区間と比較して、長期間にわたり地震が発生していない活断層の区間。地震空白域は、将来の地震発生確率が高いことを示唆する場合がある。 analysis — identifying fault segments that have not ruptured in longer than their average interval and are therefore "overdue" — relies directly on paleoseismic data. The Coachella segment of the San Andreas 断層岩盤に沿って動きが生じた破断面。断層は数ミリメートルから数千キロメートルまで様々な長さがある。地震を引き起こす主要な断層は活断層と呼ばれる。, which last ruptured around 1680 CE based on paleoseismic evidence, has an estimated recurrence interval of roughly 200 years, making it one of the most worrying seismic gaps in North America.

Paleoseismology and Probabilistic Hazard

古地震学断層トレンチ、隆起した段丘、津波堆積物といった地質学的証拠を通じて、先史時代の地震を研究する学問。地震の記録を数千年前まで遡らせる。 results feed directly into the fault source models used in 確率論的地震ハザード解析（PSHA）起こりうるすべての地震の発生源・マグニチュード・地震動レベルを考慮し、特定の揺れの水準を超える確率として結果を表現する、地震ハザードの定量評価手法。. The recurrence intervals, slip per event, and fault slip rates断層に沿った変位の平均速度で、通常は年間ミリメートル単位で測定される。すべり速度が高いほど、一般に地震発生頻度とリスクが高いことを示す。 derived from trench studies populate the input parameters for fault source characterization models such as those in the United States National Seismic Hazard Model (NSHM). Uncertainties in paleoseismic data — imprecise radiocarbon ages, ambiguous event horizons, incomplete exposure in trenches — are propagated through the hazard model using logic trees that capture the range of plausible interpretations. This formal treatment of uncertainty is essential in 確率論的地震ハザード解析（PSHA）起こりうるすべての地震の発生源・マグニチュード・地震動レベルを考慮し、特定の揺れの水準を超える確率として結果を表現する、地震ハザードの定量評価手法。, ensuring that hazard estimates honestly reflect what is and is not known about the behavior of each fault source. As 古地震学断層トレンチ、隆起した段丘、津波堆積物といった地質学的証拠を通じて、先史時代の地震を研究する学問。地震の記録を数千年前まで遡らせる。 databases grow and dating techniques improve, the quality and reliability of fault-based seismic hazard assessments continues to improve, directly contributing to safer building codes and more informed land-use planning in earthquake-prone communities.