Research Summary: Lake Sediments Provide A Natural Seismometer For Earthquakes On A Plate Boundary Fault

¹Department of Active Landscapes, GNS Science, PO Box 30-368, Lower Hutt, New Zealand

²Department of Geography, University of Otago, PO Box 56, Dunedin, New Zealand

The Alpine Fault is one of the longest and fastest slipping plate boundary transform faults on Earth and represents the largest source of seismic hazard for New Zealand’s South Island (Berryman et al., 2012; Fig, 1). Despite its high seismic hazard, little is known about the spatial extent of rupture and magnitude of earthquakes that occurred before the last Mw>8 earthquake in 1717 A.D. Traditional paleoseismic approaches that trench the surface trace of the fault to identify and date past earthquakes have not yielded data of sufficient quality to determine when, where, and how much of the fault ruptured in past earthquakes (e.g. De Pascale and Langridge, 2012; Berryman et al., 2012). We applied an alternative approach using lake sediments to date and reconstruct the spatial extent of rupture (Howarth et al., 2012; 2014). Lakes situated adjacent to active faults are sensitive recorders of seismic shaking during earthquakes (Moernaut et al., 2014; Howarth et al., 2012; Waldmann et al., 2011). By identifying deposits formed by the high intensity shaking that occurs proximal to a ruptured fault and correlating between lakes spaced along its length, we have developed a rupture history for the last five Alpine Fault earthquakes.

Figure 1: Location map. A: The Alpine Fault in South Island, New Zealand and the location of lakes Paringa and Mapourika. B,C: Lakes Mapourika and Paringa and their catchments. D: Photo of Lake Paringa looking southeast towards the Alpine Fault and the frontal ranges of the Southern Alps. (Credit: Jamie Howarth and Sean Fitzimons)

Methods

The sedimentary record of Lakes Mapourika and Paringa (Fig. 1), situated immediately adjacent to the Alpine Fault, were sampled using a Mackereth corer to retrieve 6 m sediment cores (Mackereth, 1958; Fig. 2). Cores were logged visually and their physical properties characterised using a GEOTEK™ multi sensor core logger. Core physical properties together with grain-size and organic chemistry were used to identify deposits in the cores and to aid interpretation of flow processes and sediment sources. Chronologies for the sedimentary records were established by radiocarbon (14C) dating more than 40 terrestrial leaf macrofossils using Accelerator Mass Spectrometry (AMS) (Howarth et al., 2013). The radiocarbon dates were combined with stratigraphic information using a Bayesian age modelling approach (Bronk Ramsey, 2008). The age model was used to derive the timing of subaqueous mass-wasting events and phases of increased terrestrial sediment flux to the lake basin. The precise chronology allowed the most recent events to be correlated with the last two known earthquakes on the Alpine Fault. Correlation between the last two Alpine Fault earthquakes and the lake records facilitated comparison of modelled shaking intensities and the lake deposits allowing different levels of shaking to be assigned to the shaking-related deposits.

Figure 2: Mackereth corer surfacing after taking a 6 m sediment core. (Credit: Jamie Howarth and Sean Fitzimons)

Results

In lakes Mapourika and Paringa, seismic shaking is recorded as turbidites formed by coseismic subaqueous slope failures, which are overlain by sets of hyperpycnites representing elevated fluvial sediment fluxes from earthquake-induced landslides in the lake catchments. Deposits formed by seismic shaking are separated by layered silts that represent deposition during periods of relative quiescence in the lake catchments (Fig. 3). We demonstrated seismic triggering of the subaqueous mass-wasting and widespread landsliding that formed the event deposits: 1) establishing that events were synchronous between the lakes, which are separated by more than 80 km; 2) correlating the timing and sedimentology of these events with known ruptures of the Alpine Fault; and 3) discounting other trigger mechanisms such as extreme storms. By comparing deposits and modelled shaking intensities at the lake sites for the last two Alpine Fault earthquakes, we have shown that the two responses record different levels of shaking. Strong shaking intensities (Modified Mercali Intensity (MM) >VI) destabilise slopes within lake basins causing subaqueous slope failures and deposition of turbidites. When shaking intensities exceed MM IX extensive earthquake-induced landsliding is triggered on hillslopes in lake catchments, driving dramatic increases in sediment flux that form hyperpycnite sets in the lake basins. The presence of these two types of deposits means that lake sediments can be used to distinguish between strong (MM VI) and violent (MM IX) shaking at a lake site.

Figure 3: Example of a sediment core from Lake Paringa. In-set shows depositions over an Alpine Fault seismic cycle showing a turbidite formed by coseismic subaqueous slope failures (orange), a postseismic increase in fluvial sediment flux to the lake (yellow), returning to sediments deposited during normal conditions (brown). (Credit: Jamie Howarth and Sean Fitzimons)

Conclusions

The ability to map the spatial extent of MM IX shaking provides new insights into the timing and extent of rupture for the last five earthquakes on the Alpine Fault. The Lake Mapourika and Paringa records have refined the spatial extent of rupture for the penultimate event and identified three older earthquakes that ruptured at least the Central section of the Alpine Fault, which ruptured in 1717 A.D., 1409-1389 A.D., 1107-1055 A.D. and 961-894 A.D. providing a recurrence interval of 271±73 yrs. Lakes suitable for similar paleoseismic investigation are situated along the entire length of the Alpine Fault and ongoing work seeks to determine the rupture behaviour along the length of the fault over the last 3000 years.

Full study published in Earth and Planetary Science Letters, 2014 v. 403, 340-351.

Selected References

1. Berryman, K.R., Cochran, U.A., Clark, K.J., Biasi, G. P., Langridge, R.M., and Villamor, P., 2012a, Major earthquakes occur regularly on an isolated plate boundary fault. Science, v. 336, p.1690-1693.
2. Bronk Ramsey, C.B. 2008, Deposition models for chronological records. Quaternary Science Reviews, v. 27, p. 42-60.
3. De Pascale, G.P., and Langridge R., 2012, New on-fault evidence for a great earthquake in A.D. 1717, central Alpine Fault, New Zealand. Geology, v. 40, p. 791-794.
4. Howarth, J.D., Fitzsimons, S.J., Norris, R.J., and Jacobsen, G.E. 2012. Lake sediments record cycles of sediment driven by large earthquakes on the Alpine Fault, New Zealand. Geology, v. 40, p. 1091-1094.
5. Howarth, J. D., Fitzsimons, S. J., Jacobsen, G. E., Vandergoes, M. J., and Norris, R. J., 2013, Identifying a reliable target fraction for radiocarbon dating sedimentary records from lakes. Quaternary Geochronology, v. 17 p. 68-80.
6. Howarth, J. D., Fitzsimons, S. J., Norris, R. J., and Jacobsen, G. E. 2014, Lake Sediments record high intensity shaking that provides insight into the location and rupture length of large earthquakes on the Alpine Fault, New Zealand. Earth and Planetary Science Letters, v. 403, 340-351.
7. Mackereth, F. J. H., 1958, A portable core sampler for lake deposits. Limnology and Oceanography, v. 3, p. 181-191.
8. Moernaut, J., Van Daele, M., Heirman, K., Fontijn, K., Strasser, M., Pino, M., Urrutia, R. and De Batist, M. 2014, Lacustrine turbidites as a tool for quantitative earthquake reconstruction: New evidence for a variable rupture mode in south central Chile. Journal of Geophysical Research: Solid Earth, v. 119, p. 1607-1633 2013JB010738. doi: 10.1002/2013JB010738.
9. Waldmann, N., Anselmetti, F. S., Ariztegui, D., Austin, J. J. A., Pirouz, M., Moy, C. M., and Dunbar, R., 2011, Holocene mass-wasting events in Lago Fagnano, Tierra del Fuego (54°S): implications for paleoseismicity of the Magallanes-Fagnano transform fault. Basin Research, v. 23, no. 2, p. 171-190.