T21E-01 INVITED
Laser Microprobe (U-Th)/He Thermochronology of Detrital Minerals
A persistent concern in detrital mineral geochronology is the need to obtain a representative sampling of crystallization or cooling ages in the source region. Methods with high throughput --- e.g., laser microprobe 40Ar/39Ar thermochronology of muscovite and U-Pb thermochronology of zircon --- have a distinct advantage in this regard. Both techniques have advanced to the point that the dozens of analyses necessary to obtain a representative sampling can be done quickly and with sufficiently precision for high-quality research. Datasets obtained using methods that are far more labor intensive --- e.g., single-grain (U-Th)/He and fission track dating of minerals such as zircon --- typically include many fewer analyses. Consequently, we have less confidence that the cooling age distribution in the dataset represents the cooling age distribution in the source region. Of greater concern are analytical protocols that increase the probability of non-representative sampling. One example is the practice of picking zircon grains that are inclusion-free and euhedral (or nearly so) for conventional (U-Th)/He dating. While this practice is essential for successful conventional (U-Th)/He dating, it unavoidably leads to the systematic exclusion of grains that actually may represent significant portions of the source terrain. We describe a new approach to detrital mineral (U-Th)/He thermochronology that, in principle, provides a higher- fidelity record of the source region cooling history than the conventional technique. It involves the use of an excimer laser microprobe to ablate portions of the grain interiors from detrital zircons in a polished grain mount. (Prior to analyses, the grains can be mapped using backscattered electron and cathodoluminesence imagery.) The amounts of evolved 4He are typically so small that they are best measured using a magnetic-sector mass spectrometer rather than a quadrupole mass spectrometer of the type typically used for conventional (U- Th)/He dating. U and Th concentrations can be measured in or around the same ablation pit using laser ablation, inductively coupled plasma source mass spectrometry or secondary ionization mass spectrometry. The greatest difficulty in widespread application of the technique, at present, is the lack of reliable U and Th concentration standards. In addition to making ill-formed grains amenable to analysis, the method allows a user to target specific areas within grains and thus avoid regions with inclusions or complex chemical zoning, both of which complicate conventional dating. This presentation will be a review of the current state of development of this promising technique, as well as a critical evaluation of its strengths and weaknesses compared to conventional (U-Th)/He thermochronology.
T21E-02 INVITED
Detrital dating of Asian orogenesis: insights and caveats
Technological advances over the past two decades have facilitated increasingly routine application of single- crystal dating and cosmogenic nuclide dating to studies of orogenic erosion. Both approaches commonly utilize grab samples of detrital sediment, either modern or ancient. Whereas detrital cosmogenic data are typically used to define mean erosion rates for upstream catchments, single-crystal ages are used both to discern provenance and to define lag times: interval between isotopic closure and deposition. Recent results from dating modern fluvial sediments illuminate key concepts that underpin interpretations of results from older strata: the fidelity of the detrital signal, its evolution through an orogen, its relationship to discrete source areas, and its temporal evolution. Despite the increasing availability of dates and rates for detrial grains, relatively few studies have addressed the sources of uncertainty that modulate the precision and accuracy with which detrital results should be interpreted. Such uncertainties derive not only from sampling statistics and measurement uncertainties, but also from both geomorphic sources (seasonal variation in sediment supply and source, changes in glacial cover, the impact of stochastic geomorphic events, such as landslides), as well as tectonic ones (time-dependent deformation and thermal models, particle paths through the orogen). A better understanding of the impact of these uncertainties will underpin more reliable and less speculative interpretations of future dating results from both ancient and modern detrital fluvial sediments.
T21E-03
Miocene-Pliocene range growth and basin evolution in NE Tibet, Laji-Jishi Shan and surrounding basins.
The lateral expansion and upward growth of the Tibetan Plateau remains enigmatic, despite its relevance for exploring both the geodynamics responsible for the world's largest topographic perturbation and its climatic consequences. Here we apply (U-Th)/He thermochronology, detrital zircon provenance, and magnetostratigraphy to constrain the timing and nature of uplift, erosion, and deposition along a transect that crosses the plateau's NE margin in the vicinity of the Laji-Jishi Shan. The Hualong basin has a 1.5-km-thick, coarsening-upward fluviolacustrine stratigraphic succession. High-resolution magnetostratigraphy (500 sites) yields a reasonable correlation to the magnetic timescale between 2.1 Ma - ~10.7 Ma. (Notably, this section records several short- lived geomagnetic excursions.) Hualong basin experienced a striking, 2.5-fold increase in sediment accumulation after ~8.5 Ma, presumably due to accelerated exhumation of the adjacent Laji-Jishi Shan. However, this increase could also reflect erosion of the West Qinling to the south. Given the distinctive detrital zircon U/Pb age signature of sediment derived from each range, we are using the U/Pb provenance tool to determine the relative influences of the contrasting sediment sources and their evolution. This will help discern whether the sustained increase in sedimentation is primarily due to tectonic forcing (erosion of one range) or whether intensification of the monsoon is responsible for an enhanced sediment supply regionally (erosion of multiple ranges). Additionally, new (U-Th)/He ages from the Laji-Jishi Shan exhibit two clusters: one in the early Miocene and one in the late Miocene. This may reflect two episodes of rapid cooling restricted to localized portions of the range. Erosion during the late Miocene event is a likely cause of the coeval, 2.5-fold increase in sedimentation rates. Cross-sections of the region indicate <25% Cenozoic shortening.
T21E-04
Influence of topographic evolution and faulting on detrital thermochronometer ages: Application to the Nepalese Himalaya
Detrital thermochronology is an increasingly popular technique for determining catchment-wide, long-term denudation histories in mountainous regions. However, in the rapidly eroding Himalayan front detrital age data interpretation is complex and it remains unclear how tectonic and topographic variations affect age distributions in river sediments and syn-tectonic basin deposits. In this study, we evaluate how temporal variations in topographic evolution and thrust faulting influence detrital grain-age distributions measured in modern river sediments. We utilize a 3D thermokinematic numerical model with evolving topography to investigate the effects of thrust faulting and relief change on synoptic probability density functions of various thermochronometer systems. Model inputs include fault kinematics, magnitude and timing of relief changes, thermal boundary conditions, and material properties. Model topography and tectonics are based on the Marsyandi River region in central Nepal over the last 20 My. Thrust motion is partitioned between the Main Frontal Thrust and Main Central Thrust at rates of 0.2-14.5 and 0-4.1 mm/y, respectively, generating denudation rates of 0.1-5 mm/y. Topographic relief is varied by ±50% at various times during the simulation. Results show that detrital grain-age distributions from modern rivers have a strong sensitivity to the denudation rate dictated by the fault kinematics, but are not sensitive to fault slip partitioning without relief change. However, relief changes produce age distributions that are sensitive to the fault kinematics, especially in higher-temperature thermochronometers (e.g., muscovite 40Ar/39Ar). Predicted thermochronometer age ranges also vary with relief change, generating wider age ranges with increasing relief and narrower ranges with decreasing relief. In contrast, the primary peak ages are not sensitive to relief changes at rapid exhumation rates, but secondary age peaks may develop. Taken together, these results suggest: (a) Modern river samples are sensitive to changes in relief, but this signal will be obfuscated in lag times calculated from syn-tectonic basin deposits, and (b) in rapidly eroding regions, higher temperature detrital thermochronometer samples have the greatest sensitivity to topographic evolution.
T21E-05
Did a proto-Indus fan drain to the west?
Use of detrital records to reconstruct orogenic growth history requires understanding any long-term changes in sediment routing. For the western Himalayas a proto-Indus fan (Khojak Fan) has been postulated with a drainage that would have deposited Eocene-Oligocene sediments far to the west (present-day western Pakistan and the Makran region of southern Iran) until uplift of the Murray Ridge-Kirthar fold belt in the Miocene. To test for this we conducted a detailed provenance study on Tertiary turbidites exposed in the Makran of southern Iran. The earliest of these sediments have a paleoflow that indicates they were largely sourced from the northeast, possibly from the Himalayan suture zone. Bulk petrography of the Makran sandstones show a mixed provenance, with a prominent change from dominant arc components in Eocene-Early Oligocene samples to a sharp decrease in volcanic grains and an increase in low-grade metamorphic grains in Late Oligocene-mid-Miocene samples consistent with reduced arc input and greater contribution from metamorphic basement. Detrital thermochronometric data (zircon U-Pb, fission-track and 40Ar/39Ar white mica) record a change in source between samples that span the Late Eocene - Early Oligocene (dominated by detrital age populations between 100-150 Ma) and Miocene samples (most source ages between 30-60 Ma). These age modes are consistent with arc sources as the different closure temperatures record the same ages. Contribution from Paleozoic continental basement sources does not appear to change significantly through time. The distinct change in sediment source from arc dominant in the Paleogene to arc subordinate in the Neogene coincided with Miocene uplift of the Murray-Ridge-Kirthar belt and closure of the Katawaz Ocean, a remnant of Tethys. This process closed off Cretaceous arc sources located in the circum Himalaya suture zone to leave a more localised drainage associated with the Tertiary Chagai arc. Such change is consistent with a west draining Paleo-Indus and the main peaks of zircon U-Pb ages are consistent with Karakoram-Kohistan Suture Zone sources but Cretaceous arc magmatism was fairly ubiquitous across northern Tethys.
T21E-06
Early Himalayan Erosion: Constraints to Models of Crustal Deformation, Changes in Ocean Geochemistry and Global Climate
Dating the onset of Himalayan erosion is critical to understanding crustal deformation, and the proposed link with global climate and ocean geochemistry. The most commonly quoted age of India-Asia collision is ~50 Ma, yet the record of Paleogene Himalayan erosion is scant, either absent or of low age resolution. We identify sediments shed from the rapidly exhuming southern flanks of the eastern-central Himalaya at 38 Ma, in the >1 km thick deltaic Barail Formation of the Bengal Basin, Bangladesh. This formation was previously of disputed provenance and poorly dated. New provenance data from the Barail Formation, (seismic, petrographic, geochemical, and Ar-Ar, U-Pb, ZFT, Sm-Nd and Re-Os isotopic ratios) is consistent with Himalayan, and inconsistent with Indian cratonic or Burman sources. The biostratigraphic and isotopic mineral ages date the Barail Formation as spanning Late Eocene to Early Miocene. Lag time data show that exhumation of the orogen was rapid by 38 Ma. These data 1) reduce the delay between collision, and oldest known record of substantial and rapid erosion of the Indian crust of the central-eastern Himalaya's southern flanks, from >20 Myrs to 12 Myrs, and 2) indicate that thrusting and crustal thickening had generated sufficient relief by 38 Ma to generate efficient erosion. They also explain a previously puzzling discrepancy between the timing of early hinterland metamorphism and exhumation inferred from bedrock and the lack of corresponding evidence in the erosional record. Older sediments derived from the orogen's southern flanks may be found in the future. However, we consider the Barail Formation to represent the onset of significant accelerated erosion from the southern flanks of the central-eastern Himalaya at a regionally applicable scale because 1) earlier input of significant clastic material to the Bengal Basin is inconsistent with the pre-Barail carbonate shelf depositional environment, 2) the Bengal Basin is the only obvious exit point for such detritus, 3) the strata in the Northeast Bengal Basin can be correlated over a large area of the delta in which major Himalayan-derived input initiated sometime around 40 Ma (Lindsay et al., GSAB 1991) and 4) the data are consistent with the exponential increase in accumulation rates at the start of the Oligocene in basins surrounding the India-Asia collision zone (Metivier et al, GJI 1999). Our study 1) supports orogenic models that imply delayed erosion, although the rapid exhumation is earlier than given by models of "channel flow" (Beaumont et al, Nature 2001), 2) supports the contention that Himalayan erosion resulted in increase of the marine 87Sr/86Sr ratio at ~40 Ma and 3) brings into question the proposed link between the onset of Himalayan erosion and the onset of Cenozoic global cooling.
T21E-07 INVITED
From Australia to Nepal: Detrital cosmo from the ends of the Earth
For over a decade cosmogenic nuclide (10-Be and 26-Al) concentrations measured in stream and river sediments have been used to yield an extraordinary wealth of information on how the Earth's surface is changing. We focus here on using detrital concentrations to infer erosion rates, a methodology that is widely used across diverse landscapes. There have, however, been relatively few studies systematically testing the three principal assumptions that underlie the methodology: 1) The mobile soil or regolith layer contributing sediment to the channel is well mixed; 2) Subcatchments contribute sediment in proportion to their long-term erosion rate; and 3) The mean residence time of the sediment in storage and transport is relatively short. Recent studies have shown effective interpretation of detrital signals where one or more of these assumptions did not hold using more complicated modeling. Here we present data from field sites in Australia and Nepal that span about three orders of magnitude in effective exposure age to reassess the veracity of the above assumptions. In Australia we have detrital nuclide concentrations from "source to sink" and from a well-studied soil-mantled landscape that test the first and third assumption, while in Nepal we have concentrations that may test the second and third for a landscape that is eroding up to three orders of magnitude more rapidly. These studies emphasize the remarkable applications that this methodology enables and caution against assuming that the simplest assumptions always hold.
T21E-08
Deformation of Isochrones and Implications for Detrital Thermochronology
We demonstrate the remedy for a key assumption in the use of detrital thermochronology to trace sediment bedrock sources by correcting bedrock cooling age distributions to account for deformation. In the south-central European Alps, the western Sierra Nevada, and Icicle Creek, Washington State, isochrones, or planes of equal cooling age, are ubiquitously sheared away from horizontal by as much as sixty degrees between origin at the closure isotherm and surface exposure. Established techniques based on probability-density functions (PDFs) fail in these localities because shearing compromises age-elevation relationships, thereby precluding the accurate prediction of age PDFs. This pitfall is evidently common to the flanks of orogens, especially those with wedge-type geometries. However, modeling of the deformed isochrones allows the production of just such predicted distributions. In this study, the modeled, best-fit isochrones account for both tilting and the possibility of a change in exhumation rate. Comparison of predicted with observed PDFs provides an estimate of which areas of the basin contribute sediment--effectively the relative erosion rates throughout the basin. Absolute erosion rates depend on the transport time through the basin's hillslopes and streams. In the Alps, we show agreement among this new detrital measure of erosion rates and other geologically short-term measures, including cosmogenic dates and leveling data.