U42A-01 INVITED
Short-Term Geomagnetic Variability as a Tool for Determining Precise Global Chronologies: Examples From the North Atlantic
Short-term variations in the strength and direction of the geomagnetic field have the potential to provide Quaternary chronologies with millennial and sub-millennial resolution. The variations, particularly in the form of large directional swings such as geomagnetic excursions and relative paleointensity highs and lows, can typically be correlated over large regions and, in many cases, worldwide. These attributes, in addition to the relative ease and speed with which high-resolution measurements can be made, provide a means for constructing chronologies that can be used at sites around the globe. When the paleomagnetic records are tied to a variety of climate and environmental proxies\textemdash stable isotope, cosmogenic isotope, lithology, sediment color, rock magnetic, and other records\textemdash they become a powerful tool for examining the synchroneity of climatic events. Besides being a passive chronologic tool, however, the geomagnetic field may also play a role in climate change or be influenced by such change. Paleomagnetic records from across the North Atlantic reveal the synchroneity of the geomagnetic variability across the region, including the occurrence of multiple geomagnetic excursions during the Brunhes chron (0-780,000 years). Intriguing relationships between the geomagnetic variability and both climate change and orbital forcing suggest a causal link.
U42A-02 INVITED
Calibration of a Pleistocene Geomagnetic Instability Time Scale (GITS) using 40Ar/39Ar-dated lavas
Advances in measuring paleomagnetic intensity recorded by marine sediments, and 40Ar/39Ar dating of paleomagnetic directional recordings in lava flows, offer a means of calibrating a global magnetostratigraphy for the last 2 m.y. This involves moving beyond the classic geomagnetic polarity time scale (GPTS) and resolving not only the undisputed polarity reversals, but also the many short-lived geomagnetic "events" or cryptochrons thought to signal brief periods of geodynamo instability. Many short events are distinguished as intensity minima in global sediment records (SINT-800; GLOPIS-75) that are dated by astrochronology. Thus, when the degree of stability of the geodynamo is considered, rather than lengths of polarity intervals, an alternative approach to the GPTS is appropriate. We are developing a Geomagnetic Instability Time Scale (GITS) via 40Ar/39Ar dating of transitionally-magnetized lava flows younger than 2 Ma. As an example, the Laschamp event--expressed as a sharp intensity minimum in the GLOPIS-75 sediment stack--was dated by matching O-isotope variations in North Atlantic sediments to those recorded in annually counted layers of the GISP2 ice core. Matching 14C ages from the sediments to specific varves in the ice core shows the paleointensity minimum to span 1500 yr between 41 and 39 ka. 40Ar/39Ar and unspiked K-Ar dating of two basaltic lava flows that record the event at Laschamps, France yield an age of 40.4±1.1 ka (± 2 sigma, analytical uncertainty). Thus, despite systematic uncertainty in the 40K decay constant, both the accuracy and precision of the K-Ar clock can be remarkably good, i.e., better than 2% for the Pleistocene. Intercomparison of several 40Ar/39Ar-dated geomagnetic events, including the Matuyama-Brunhes polarity reversal (776 ± 2 ka), Big Lost event (579 ± 6 ka) and Pringle Falls/Albuquerque event (211 ± 11 ka) implies either that: 1) the astrochronology-based age models used for the SINT-800 paleointensity stack are inaccurate between 150 and 600 ka, or 2) the excursional directions recorded in the dated lava flows do not correlate with global lows in paleointensity.
U42A-03
A new Counted Greenland ice Core Chronology (GICC05)
During a five year grant the Ice and Climate group at Niels Bohr Institute, University of Copenhagen is performing annual layer counts in the Dye-3, GRIP and NGRIP ice cores to obtain a Greenland ice core chronology on past climatic and environmental events. The counting is based on seasonal variations in stable isotopes, in chemical impurities and dust and in visual stratigraphy. Using ash layers and acid layers from major volcanic eruptions as primary stratigraphic fixpoints and acid layers from smaller volcanoes as secondary fixpoints, the three ice core records have been stratigraphically "inter locked". This has allowed for a compilation of the three chronologies into one. I will present the new Greenland Ice core Chronology which until now extends back to 40,000 years B.P. This chronology provides dates on a number of abrupt climate changes, on a number of volcanic tephra layers and on variations in atmospheric concentrations of 10Beryllium. It will thus be possible to link the Greenland ice core chronology to marine sediment records, to terrestrial records, to dendrochronologies and the Carbon-14 calibration curve. Although a certain counting uncertainty remains and the dating of deep ice strata will be influenced by this, the new chronology offers very precise differential dating possibilities. It is possible to track the abrupt climate changes on a year by year basis and the age difference between such transitions and e.g. a volcanic eruption from the same period can be determined with uncertainties of a few years. Since a row of different parameters have been measured at high resolution in the NGRIP ice core it has been possible to obtain a picture of the dynamics across abrupt climate changes. The data allows us to determine the termination of the Younger Dryas within less than 5 years.
U42A-04
Age of the Xalnene Ash, Central Mexico and Archeological Implications
Human footprints ~40 ka old have been reported from the Toloquilla quarry near Valsequillo Reservoir, ca. 15 km south of the city of Puebla in central Mexico (http://www.mexicanfootprints.co.uk/default.htm). If correct, this would be important evidence for early peopling of the Americas. The indentations interpreted as footprints and other ichnofossils occur on the surface of an indurated basaltic lapilli tuff within a several meter thick sequence of thinly bedded (1-10 cm) tuffs of similar character, lacking paleosols, erosional features or interlayered sediments, informally known as the Xalnene ash. A sample was collected at 18°55.402' N latitude and 098°09.375' W longitude from the surface on which the purported footprints occur. Lapilli were separated and analyzed by incremental heating 40Ar/39Ar methods, yielding 9 indistinguishable plateau ages averaging 1.30 ±0.03 Ma (2σ) for single lapilli (N=6) and multiple lapilli (N=3) subsamples. Though some minor discordance (presumably due to 39Ar recoil) is manifest in 5 of the age spectra, all plateaux comprise >60% of the 39Ar released and 4 or more consecutive steps. Paleomagnetic data from azimuthally unoriented bulk samples of 11.25 cm3 reveal a reverse polarity (I = -32.1°) thermoremanent component carried by titanomagnetite and a normal polarity component carried by goethite. Measurements on individual matrix-free lapilli lack the goethite component, which is presumed to be associated with the clay-rich cement. Consistency of the reverse component implies deposition of the lapilli at supra-Curie temperatures, with no postdepositional reworking. Reverse polarity is consistent with deposition during chron C1r.2r (1.77 to 1.07 Ma) as indicated by the 40Ar/39Ar data. If the features observed on the tuff are indeed footprints, their 1.3 Ma antiquity would be truly remarkable, predating by far any other evidence for human presence in the Americas and in fact predating the evolutionary emergence of Homo sapiens (in Africa) by more than 1 Ma. We conclude that the identification of these features as syn-depositional human footprints is likely erroneous.
U42A-05 INVITED
A Bayesian Approach to Correlation of Paleoearthquakes on a Fault
We have developed an approach to paleoearthquake correlation for sites on the southern San Andreas fault in California. Paleoseismic sites are spaced kilometers to multiple tens of kilometers apart along the fault. One cannot normally say from paleoseismic dating alone whether a surface rupture spans multiple sites or whether the individual observations are from smaller earthquake ruptures that are merely close in time. Seismic hazard and thus societal consequences depend both on the magnitude and frequency of ground-rupturing earthquakes, so if ruptures correlate, the seismic hazard is weighted toward fewer, larger earthquakes, and if they do not, the hazard is caused by more numerous, somewhat smaller earthquakes. Our approach produces probability of correlation estimates by using additional physical measurements available from the paleoseismic investigation. The most powerful is the measurement of rupture displacement, d$_{obs}$, at a site. In outline the idea is that large point displacements generally go with longer ruptures, and longer ruptures have an increased likelihood of spanning from one site to the next. Mapped historical ruptures provide empirical relationships between average slip of a rupture, rupture length, and earthquake magnitude. Assessing the probability of correlation would be straightworward except that rupture displacement can vary along its length by up to a factor of 3 or so from its average, depending on geologic and possibly antecedent slip conditions. Especially for strike-slip faults one cannot say whether d$_{obs}$ is from a peak for a smaller earthquake, or a low point from a large earthquake. However, the degree of variability of slip for small and large ruptures is usefully similar when normalized by the average slip for the rupture. This provides the basis for a Bayesian inversion for probabilities of rupture length and earthquake magnitude given the point measurement of displacement, p(L$|$d$_{obs}$) and p(M$|$d$_{obs}$), respectively. Geometric considerations are used to form a simple probability of correlation given length, p(correl$|$L). With p(L$|$d$_{obs}$) this leads to the probability of correlation given a point observation of displacement, p(correl$|$d$_{obs}$). Correlation probabilities decline with increasing site separation and decreasing observed displacement. For the southern San Andreas fault, large site spacings and the relatively small number of slip measurements are leading challenges to a more complete rupture history of the fault.
U42A-06
Unraveling the Earthquake History of the Denali Fault System, Alaska: Filling a Blank Canvas With Paleoearthquakes
Developing accurate rupture histories of long, high-slip-rate strike-slip faults is is especially challenging where recurrence is relatively short (hundreds of years), adjacent segments may fail within decades of each other, and uncertainties in dating can be as large as, or larger than, the time between events. The Denali Fault system (DFS) is the major active structure of interior Alaska, but received little study since pioneering fault investigations in the early 1970s. Until the summer of 2003 essentially no data existed on the timing or spatial distribution of past ruptures on the DFS. This changed with the occurrence of the M7.9 2002 Denali fault earthquake, which has been a catalyst for present paleoseismic investigations. It provided a well-constrained rupture length and slip distribution. Strike-slip faulting occurred along 290 km of the Denali and Totschunda faults, leaving unruptured ?140km of the eastern Denali fault, ?180 km of the western Denali fault, and ?70 km of the eastern Totschunda fault. The DFS presents us with a blank canvas on which to fill a chronology of past earthquakes using modern paleoseismic techniques. Aware of correlation issues with potentially closely-timed earthquakes we have a) investigated 11 paleoseismic sites that allow a variety of dating techniques, b) measured paleo offsets, which provide insight into magnitude and rupture length of past events, at 18 locations, and c) developed late Pleistocene and Holocene slip rates using exposure age dating to constrain long-term fault behavior models. We are in the process of: 1) radiocarbon-dating peats involved in faulting and liquefaction, and especially short-lived forest floor vegetation that includes outer rings of trees, spruce needles, and blueberry leaves killed and buried during paleoearthquakes; 2) supporting development of a 700-900 year tree-ring time-series for precise dating of trees used in event timing; 3) employing Pb 210 for constraining the youngest ruptures in sag ponds on the eastern and western Denali fault; and 4) using volcanic ashes in trenches for dating and correlation. Initial results are: 1) Large earthquakes occurred along the 2002 rupture section 350-700 yrb02 (2-sigma, calendar-corrected, years before 2002) with offsets about the same as 2002. The Denali penultimate rupture appears younger (350-570 yrb02) than the Totschunda (580-700 yrb02); 2) The western Denali fault is geomorphically fresh, its MRE likely occurred within the past 250 years, the penultimate event occurred 570-680 yrb02, and slip in each event was 4m; 3) The eastern Denali MRE post-dates peat dated at 550-680 yrb02, is younger than the penultimate Totschunda event, and could be part of the penultimate Denali fault rupture or a separate earthquake; 4) A 120-km section of the Denali fault between tNenana glacier and the Delta River may be a zone of overlap for large events and/or capable of producing smaller earthquakes; its western part has fresh scarps with small (1m) offsets. 2004/2005 field observations show there are longer datable records, with 4-5 events recorded in trenches on the eastern Denali fault and the west end of the 2002 rupture, 2-3 events on the western part of the fault in Denali National Park, and 3-4 events on the Totschunda fault. These and extensive datable material provide the basis to define the paleoseismic history of DFS earthquake ruptures through multiple and complete earthquake cycles.
U42A-07
A Composite Chronology of Earthquakes From the Bidart fan Paleoseismic Site, San Andreas Fault, California
Chronologies of earthquakes spanning at least ten ruptures at multiple sites are required for developing robust models of fault behavior and forecasts of future earthquakes. Such a long chronology can be obtained by placing multiple trenches across the San Andreas fault at the Bidart alluvial fan paleoseismic site in the Carrizo Plain to capture the spatio-temporal record of earthquakes created by the interplay of surface rupture and spatially varying deposition. Exposures from one trench reveal evidence of at least 6 and probably 7 earthquakes since 3000 BP. Evidence of 7 earthquakes since 2200 BP has been interpreted from exposures in 3 other trenches. Analysis of exposures from two new trenches is in progress. Excavations reveal alternating sequences of depositional preservation and gaps in the record of earthquakes. The "gaps" are massive featureless zones caused by bioturbation of the fan surface while that portion of the fan was depositionally inactive. When the depositional record of 4 trenches is combined, it yields a composite chronology of at least10 surface ruptures over the last 3000 years, for a minimum average recurrence interval of 300 years if the most recent event exposed in all trenches is assumed to be the 1857 Fort Tejon earthquake. So far, the uncertainty in dates of pre-1857 ruptures ranges from decades to millennia, and at least 5 of the 10 recognized earthquakes are obscured by depositional gaps at one of the trench sites. Therefore, synchroneity of ruptures at different trench sites is difficult to establish, and there is the possibility that the existing record contains more than 10 earthquakes and/or additional ruptures may have occurred that are not preserved by deposition.