Dating and characterizing late holocene earthquakes using paleomagnetics

In this thesis I apply paleomagnetic techniques to paleoseismological problems. I investigate the use of secular-variation magnetostratigraphy to date prehistoric earthquakes; I identify liquefaction remanent magnetization (LRM), and I quantify coseismic deformation within a fault zone by measuring the rotation of paleomagnetic vectors.

In Chapter 2 I construct a secular-variation reference curve for southern California. For this curve I measure three new well-constrained paleomagnetic directions: two from the Pallett Creek paleoseismological site at A.D. 1397-1480 and A.D. 1465-1495, and one from Panum Crater at A.D. 1325-1365. To these three directions I add the best nine data points from the Sternberg secular-variation curve, five data points from Champion, and one point from the A.D. 1480 eruption of Mt. St. Helens. I derive the error due to the non-dipole field that is added to these data by the geographical correction to southern California. Combining these yields a secular variation curve for southern California covering the period A.D. 670 to 1910, with the best coverage in the range A.D. 1064 to 1505.

In Chapter 3 I apply this curve to a problem in southern California. Two paleoseismological sites in the Salton trough of southern California have sediments deposited by prehistoric Lake Cahuilla. At the Salt Creek site I sampled sediments from three different lakes, and at the Indio site I sampled sediments from four different lakes. Based upon the coinciding paleomagnetic directions I correlate the oldest lake sampled at Salt Creek with the oldest lake sampled at Indio. Furthermore, the penultimate lake at Indio does not appear to be present at Salt Creek. Using the secular variation curve I can assign the lakes at Salt Creek to broad age ranges of A.D. 800 to 1100, A.D. 1100 to 1300, and A.D. 1300 to 1500. This example demonstrates the large uncertainties in the secular variation curve and the need to construct curves from a limited geographical area.

Chapter 4 demonstrates that seismically induced liquefaction can cause resetting of detrital remanent magnetization and acquisition of a liquefaction remanent magnetization (LRM). I sampled three different liquefaction features, a sandbody formed in the Elsinore fault zone, diapirs from sediments of Mono Lake, and a sandblow in these same sediments. In every case the liquefaction features showed stable magnetization despite substantial physical disruption. In addition, in the case of the sandblow and the sandbody, the intensity of the natural remanent magnetization increased by up to an order of magnitude.

In Chapter 5 I apply paleomagnetics to measuring the tectonic rotations in a 52 meter long transect across the San Andreas fault zone at the Pallett Creek paleoseismological site. This site has presented a significant problem because the brittle long-term average slip-rate across the fault is significantly less than the slip-rate from other nearby sites. I find sections adjacent to the fault with tectonic rotations of up to 30°. If interpreted as block rotations, the non-brittle offset was 14.0+2.8, -2.1 meters in the last three earthquakes and 8.5+1.0, -0.9 meters in the last two. Combined with the brittle offset in these events, the last three events all had about 6 meters of total fault offset, even though the intervals between them were markedly different.

In Appendix 1 I present a detailed description of my standard sampling and demagnetization procedure.

In Appendix 2 I present a detailed discussion of the study at Panum Crater that yielded the well-constrained paleomagnetic direction for use in developing secular variation curve in Chapter 2. In addition, from sampling two distinctly different clast types in a block-and-ash flow deposit from Panum Crater, I find that this flow had a complex emplacement and cooling history. Angular, glassy "lithic" blocks were emplaced at temperatures above 600° C. Some of these had cooled nearly completely, whereas others had cooled only to 450° C, when settling in the flow rotated the blocks slightly. The partially cooled blocks then finished cooling without further settling. Highly vesicular, breadcrusted pumiceous clasts had not yet cooled to 600° C at the time of these rotations, because they show a stable, well clustered, unidirectional magnetic vector.

Late Holocene environmental variability in the upper San Francisco Estuary as reconstructed from tidal marsh sediments

EXTRACT (SEE PDF FOR FULL ABSTRACT): Tidal marsh sediments collected from Browns Island in the lower Sacramento/San Joaquin Delta, California, are used to reconstruct environmental variability over the past 6.8 ka. Calibrated radiocarbon dates provide chronostratigraphic control. Trace metal analyses, grain-size variability, organic content, and macrofossils are used to define short- and long-term variations in relative salinity and inundation frequency. Aggradation began in subtidal fresh water conditions about 6.8 ka. Subtidal aggradation of clayey silts continued until about 6.3 ka, when conditions shifted toward a lower intertidal brackish marsh environment. By 5.1 ka, a brackish marsh plain had evolved, with surface water freshening after 4.1 ka. Conditions returned to brackish similar to the present after 2.3 ka.

Late Holocene linkages between decade-century scale climate variability and productivity at Lake Tanganyika, Africa

Microlaminated sediment cores from the Kalya slope region of Lake Tanganyika provide a near-annually resolved paleoclimate record between similar to 2,840 and 1,420 cal. yr B.P. demonstrating strong linkages between climate variability and lacustrine productivity. Laminae couplets comprise dark, terrigenous-dominated half couplets, interpreted as low density underflows deposited from riverine sources during the rainy season, alternating with light, planktonic diatomaceous ooze, with little terrigenous component, interpreted as windy/dry season deposits. Laminated portions of the studied cores consist of conspicuous dark and light colored bundles of laminae couplets. Light and dark bundles alternate at decadal time scales. Within dark bundles, both light and dark half couplets are significantly thinner than within light bundles, implying slower sediment accumulation rates during both seasons over those intervals.

Late Holocene climatic changes in Tierra del Fuego based on multiproxy analyses of peat deposits

A ca. 1400-yr record from a raised bog in Isla Grande, Tierra del Fuego, Argentina, registers climate fluctuations, including a Medieval Warm Period, although evidence for the 'Little Ice Age' is less clear. Changes in temperature and/or precipitation were inferred from plant macrofossils, pollen, fungal spores, testate amebae, and peat humification. The chronology was established using a C-14 wiggle-matching technique that provides improved age control for at least part of the record compared to other sites. These new data are presented and compared with other lines of evidence from the Southern and Northern Hemispheres. A period of low local water tables occurred in the bog between A.D. 960-1020, which may correspond to the Medieval Warm Period date range of A.D. 950-1045 generated from Northern Hemisphere tree-ring data. A period of cooler and/or wetter conditions was detected between ca. A.D. 1030 and I 100 and a later period of cooler/wetter conditions estimated at ca. cal A.D. 1800-1930, which may correspond to a cooling episode inferred from Law Dome, Antarctica. (C) 2004 University of Washington. All rights reserved.

Mid- to late-Holocene climate variability and anthropogenic impacts: multi-proxy evidence from Lake Bliden, Denmark

We conducted multi-proxy geochemical analyses (including measurements of organic carbon, nitrogen and sulphur stable isotope composition, and carbonate carbon and oxygen isotope composition) on a 13.5 m sediment core from Lake Bliden, Denmark, which provide a record of shifting hydrological conditions for the past 6,700 years. The early part of the stratigraphic record (6,700-5,740 cal year BP) was wet, based on delta O-18(carb) and lithology, and corresponds to the Holocene Thermal Maximum. Shifts in primarily delta O-18(carb) indicate dry conditions prevailed from 5,740 to 2,800 cal year BP, although this was interrupted by very wet conditions from 5,300 to 5,150, 4,300 to 4,050 and 3,700 to 3,450 cal year BP. The timing of the latter two moist intervals is consistent with other Scandinavian paleoclimatic records. Dry conditions at Lake Bliden between 3,450 and 2,800 cal year BP is consistent with other paleolimnological records from southern Sweden but contrasts with records in central Sweden, possibly suggesting a more northerly trajectory of prevailing westerlies carrying moisture from the North Atlantic at this time. Overall, fluctuating moisture conditions at Lake Bliden appear to be strongly linked to changing sea surface temperatures in the Greenland, Iceland and Norwegian seas. After 2,800 cal year BP, sedimentology, magnetic susceptibility, delta C-13(ORG), delta C-13(carb) and delta O-18(carb) indicate a major reduction on water level, which caused the depositional setting at the coring site to shift from the profundal to littoral zone. The Roman Warm Period (2,200-1,500 cal year BP) appears dry based on enriched delta O-18(carb) values. Possible effects of human disturbance in the watershed after 820 cal year BP complicate attempts to interpret the stratigraphic record although tentative interpretation of the delta O-18(carb), magnetic susceptibility, delta C-13(ORG), delta C-13(carb) and delta O-18(carb) records suggest that the Medieval Warm Period was dry and the Little Ice Age was wet.

Dinoflagellate cyst-based reconstructions of mid to late Holocene winter sea-surface temperature and productivity from an anoxic fjord in the NE Pacific Ocean

Published contemporary dinoflagellate distributional data from the NE Pacific margin and estuarine environments (n = 136) were re-analyzed using Canonical Correspondence Analysis (CCA) and partial Canonical Correspondence Analysis (pCCA). These analyses illustrated the dominant controls of winter temperature and productivity on the distribution of dinoflagellate cysts in this region. Dinoflagellate cyst-based predictive models for winter temperature and productivity were developed from the contemporary distributional data using the modern analogue technique and applied to subfossil data from two mid to late Holocene (~5500 calendar years before present–present) cores; TUL99B03 and TUL99B11, collected from Effingham Inlet, a 15 km long anoxic fjord located on the southwest coast of Vancouver Island that directly opens to the Pacific Ocean through Barkley Sound. Sedimentation within these basins largely comprises annually deposited laminated couplets, each made up of a winter deposited terrigenous layer and spring to fall deposited diatomaceous layer. The Effingham Inlet dinoflagellate cyst record provides evidence of a mid-Holocene gradual decline in winter SST, ending with the initiation of neoglacial advances in the region by ~3500 cal BP. A reconstructed Late Holocene increase in winter SST was initiated by a weakening of the California Current, which would have resulted in a warmer central gyre and more El Niño-like conditions.