T31B-1289 0800h
Erosion of the Alborz Mountains Constrained by River Gauging
The Alborz Mountains of north Iran are an active fold and thrust belt within the Arabia-Eurasia collision zone, located at the south edge of South Caspian Basin. About 36$%$ of the continental convergence in this region is accommodated in the mountain belt. There is a strong climate asymmetry across the range; the north flank receives ten times more precipitation than the arid south flank, making this area ideal site to study climate control on erosion and crustal deformation. Suspended sediment records have been used to calculate erosion rates in the Alborz Mountains. 286 hydrometric stations in the range have an average period of record of 12 years of suspended load and water discharge. Of these, 144 stations have reported $>$100 suspended load measurements. For these stations, we have calculated erosion rates using mean sediment loads, converted for non-uniform sampling intervals, seasonal rating curves of discharge-suspended load and time series analysis. Sediment accumulation rates in reservoirs are employed to validate our decadal erosion rate estimates, and to constrain bedload contribution. The total suspended sediment discharge from the mountain belt is 28.5 $\times$ 10$^{6}$ t.y$^{-1}$, giving an average erosion rate of 0.16mm.y$^{-1}$ for the Alborz Mountains. Erosion is not uniformly distributed; 40.6$%$ of sediment discharge comes from just 5.9$%$ of the area. Erosion rates are highest in the east and west and lower in the central block. Notably, no significant contrast exists between erosion rates in north and south flank. Our findings imply that climate has not been an important control on decadal erosion in the Alborz Mountains; the average annual sediment yield from the wet north flank and dry south flank are 7.8 $\times$ 10$^{6}$ t.y$^{-1}$ and 20.7 $\times$ 10$^{6}$ t.y$^{-1}$ respectively. We attribute localized, fast erosion to recent seismicity.
T31B-1290 0800h
Reach-Scale Adjustments in Alluvial Channel Morphology in Response to Active Folding
Recent interest in the role of channel adjustment in response to tectonic or climatic perturbations highlights the need for field-based calibration of width-scaling relationships for channels in both alluvial and bedrock rivers. Because such adjustments may serve as a primary means for focusing erosive power in fluvial systems, a quantitative understanding of morphologic variation as a function of differential tectonic uplift provides valuable insight into fluvial response and behavior within actively deforming landscapes. In order to document changes in channel morphology in response to tectonic forcing, we have conducted detailed surveys of formerly antecedent outwash channels preserved as wind gaps over actively growing folds along the Ostler fault zone of southern New Zealand. Surveying focused on channel width and the depth of incision for over 30 cross profiles on each channel examined. Results show marked narrowing of channel widths over the proximal flanks of anticlinal uplifts toward minimum values well before reaching the uplift crest. We compare downstream variations in channel morphology with contemporaneous spatial and temporal changes in channel incision. In the context of specific stream-power models for fluvial erosion, we then assess the relative contribution of channel width and slope adjustments in enabling incision to keep pace with uplift for the life of the antecedent stream. Despite difficulties in reconstructing channel slope in deformed wind gaps, comparison with both modern unperturbed drainage gradients upstream of the fault zone and currently oversteepened stream gradients over the active hangingwall anticline indicates that adjustments in slope also play an important role in focusing erosive power through the growing structure and in allowing channels to maintain their course over active folds. In each case, our results highlight the interdependence of variations in channel morphology and imposed changes in slope and elevation through differential uplift. Such data also provides new field constraints on existing width-scaling relationships for alluvial rivers in actively deforming terrains.
T31B-1291 0800h
A New Approach to Scaling Channel Width in Bedrock Rivers and its Implications for Modeling Fluvial Incision
The fundamental role of bedrock channel incision on the evolution of mountainous topography has become a central concept in tectonic geomorphology over the past decade. During this time the stream power model of bedrock river incision has immerged as a valuable tool for exploring the dynamics of bedrock river incision in time and space. In most stream power analyses, river channel width--a necessary ingredient for calculating power or shear stress per unit of bed area--is assumed to scale solely with discharge. However, recent field-based studies provide evidence for the alternative view that channel width varies locally, much like channel slope does, in association with spatial changes in rock uplift rate and erodibility. This suggests that simple scaling relations between width and discharge, and hence estimates of stream power, don't apply in regions where rock uplift and erodibility vary spatially. It also highlights the need for an alternative to the traditional assumptions of hydraulic geometry to further investigation of the coupling between bedrock river incision and tectonic processes. Based on Manning's equation, basic mass conservation principles, and an assumption of self-similarity for channel cross sections, we present a new relation for scaling the steady-state width of bedrock river channels as a function of discharge (Q), channel slope (S), and roughness (K$_{s}$): W \propto Q$^{3/8}$S$^{-3/16}$K$_{s}$$^{1/16}$. In longitudinally simple, uniform-concavity rivers from the King Range in coastal Northern California, the model emulates traditional width-discharge relations that scale channel width with the square root of discharge. More significantly, our relation describes river width trends for the Yarlung Tsangpo in SE Tibet and the Wenatchee River in the Washington Cascades, both rivers that narrow considerably as they incise terrain with spatially varied rock uplift rates and/or lithology. We suggest that much of observed channel width variability is a simple consequence of the tendency for water to flow faster in steeper reaches and therefore maintain smaller channel cross sections. We demonstrate that using conventional scaling relations for bedrock channel width can significantly underestimate stream power variability in bedrock channels, and that our model improves estimates of spatial patterns of bedrock incision rates.
T31B-1292 0800h
Testing models of fluvial incision under conditions of differential rock uplift
It is now well recognized that fluvial incision into bedrock exerts a primary control on the topographic evolution of mountain ranges and strongly modulates the exhumational and geodynamic evolution of active orogens. Despite recent advances in the theoretical development of fluvial incision models and comparison to experimental results, significant uncertainty remains regarding model form and parameterization. In particular, relatively few tests of fluvial incision models have been conducted against data sets collected at reach-averaged length scales and over geologic time scales; recent attempts have met with only limited success. Here we test a suite of bedrock incision models under conditions of differential rock uplift rate. Although theoretical considerations suggest that channel profiles adjusted to spatially invariant rock uplift may not be diagnostic of model form (Whipple and Tucker, 2002), channels adjusted to strong spatial gradients in rock uplift posses several characteristics, including variation in the direction of tectonic forcing relative to channel flow, which may help discriminate between competing models. We examine a suite of channels developed across a growing fold in the Himalayan foreland. We combine topographic data of channel gradients (derived from high-resolution DEMs and topographic maps) and measurements of channel width (derived from aerial photography) with channel incision rates (Lav\'{e} and Avouac, 2000) and sediment flux to test against the predictions of various fluvial incision models. Preliminary results suggest that 1) channel gradients increase linearly with rock uplift/incision rate, 2) the magnitude of gradient increase is similar on channels carrying significantly different sediment loads, and 3) channel width is invariant with incision rate across a range from 5-14 mm/yr at the center of the fold. Although two end-member models (detachment and transport-limited) successfully reproduce the linear relationship observed between channel gradient and incision rate, the apparent lack of dependence of channel gradient on sediment flux is suggestive of a local control on channel gradients in this landscape.
T31B-1293 0800h
The Role of Bed Coverage in Reducing the Area Dependence of Fluvial Incision: Implications for the Maximum Relief of Mountains.
Analytical models of the coupled behavior between fluvial incision and wedge tectonics predict that the dependence of fluvial incision on drainage area is an important factor for understanding how climate (through precipitation) affects the height of active mountain belts. These models use a stream-power formulation $I=KA^mS^n$ to represent fluvial incision, but this equation is only an approximation of the many processes that control fluvial incision. Although stream power is widely applied, it is not well tested. Using first principles and scaling laws for channel flow, the value of $m$ in the stream-power equation is derived to be between 0.3 and 1.0. Our contribution here is to consider empirical estimates of $m$ and $n$, which we will refer to as $\mu$ and $\nu$. These estimates are made by fitting a power-law function to slope, area, and incision rate data and are independent from assumptions about the incision process. We use data from both natural settings (the Clearwater River, Washington State, USA, and the Lachlan River, SE Australia) and synthetic examples obtained using the CHILD landscape-evolution model. In both of the natural settings, all estimates of $\mu$ are less than 0.2 and sometimes negative. Likewise for the simulated landscapes, when sediment load is included in the incision model and uplift varies spatially, power-law estimates produce a wide range of $\mu$ values and also include negative values. The CHILD model demonstrates that the estimated $\mu$ is quite different from the value of $m$ used in the model calculation. The reason seems to be that the bed-coverage effect is correlated with drainage area for these examples. The results raise the question of what parameter values should be used in coupled analytical models. If the empirical estimates are relevant, then the low-area dependence of incision rate (as indicated by estimates of $\mu$) suggests that climate plays a very weak role in controlling the height of mountains.
T31B-1294 0800h
Erosional controls on mountainous topography, Southern Alps, New Zealand
The Southern Alps, New Zealand are proposed to be in steady state such that in the west high rates of uplift ($>$10 mm/yr) are balanced by similar rates of erosion. Erosional processes driven by a strong precipitation gradient (15 to $<$1 m/yr) across the range creates three erosional domains: a fluvially dissected, vegetated landscape on the wet, western side of the Southern Alps, an axial zone of U-shaped glacial valleys, and a broad, poorly dissected landscape with braided rivers and scree-mantled slopes in the low rainfall eastern slopes. The influence of uplift and precipitation on the spatial styles of erosion across the mountain range remains unclear. Here we quantified spatial patterns of erosion by measuring the elevation, slope, drainage density, and local relief in an NW-SE oriented, 80 km long transect across the Southern Alps from the range bounding Alpine Fault. We coupled these analyses with erosion rate estimates from a variety of sources in attempt to understand temporal effects. In the west, there is a constant $\sim$70% slope at all elevations and a high drainage density caused by fluvial dissection and threshold landsliding. Along the range axis, lower slope angles ($<$30%) are common at lower elevations and steep ($>$80%) slope angles dominate at high elevations, and drainage density decreases reflecting a change to a glacial landscape. Eastward, at 35-40 km SE of the Alpine Fault the distribution of slope angles becomes bimodal, drainage density is low, with a significant proportion of the landscape at 200-500 m elevation, reflecting a change to a depositional landscape with braided streams and scree-mantled slopes. The boundaries between these three erosional domains correspond with major boundaries in precipitation and elevation in the landscape, suggesting that both parameters serve as fundamental controls on erosional process. The domain boundaries are not reflected in the distribution of local relief, which is constant across the western and axial parts of the range and decreases as valley width increases to the east. This pattern reflects the inherited signature of glacial valleys formed during the LGM. Postglacial erosion has dissected these valleys in the west and infilled them in the east, yet slopes in both areas continue to adjust their form. This analysis emphasizes the diverse nature of process-form linkages in an active orogen and suggests that much of the Southern Alps terrain varies over glacial-interglacial timescales.
T31B-1295 0800h
Impact of ice sheet glaciation on the late Neogene exhumation of East Greenland
Rapid late Neogene denudation of landmasses around the North Atlantic appears to have coincided with the transition to the wetter and ultimately glacial climate of the Quaternary. Denudation is a powerful way of altering global climate because erosion impacts the consumption of atmospheric carbon dioxide by weathering; hence, the onset of large-scale northern-hemisphere glaciation may have contributed significantly to global cooling during the late Cenozoic by enhancing global silicate weathering. In East Greenland, thermal histories obtained using apatite fission track thermochronometry (AFTT) suggest exhumation of up to 2 km during the last 20 Ma; however, the timing and magnitude of this event are poorly constrained because cooling appears mainly to have occurred below the AFTT partial annealing zone. We report initial apatite (U-Th)/He thermochronometry results aimed at rigorously constraining the impact of ice-sheet glaciation on late Neogene exhumation in East Greenland. New data are presented from two vertical profiles collected in steep-sided glacial fjords in the Kong Oscar Fjord region of East Greenland by the Cambridge Arctic Shelf Programme. Forward modelling using best-fit cooling histories from AFTT for these samples predicts He ages of 9-10 Ma; however, measured He ages cluster around 40-50 Ma. In addition, He ages appear to correlate with mean long-wavelength topography, with the oldest ages being from the valley bottom of a minor tributary fjord. Initial results therefore suggest that: (1) denudation was earlier, more protracted and/or of lower magnitude than suggested by AFTT thermal histories; and (2) erosion by ice-sheet glaciation has been largely confined to the inception of the present-day fjord system. Further apatite (U-Th)/He thermochronology together with in situ cosmogenic nuclide analysis will be aimed at a better understanding of the controls on ice-sheet erosion and its significance in late Neogene exhumation and global climate change.
T31B-1296 0800h
Rates of Glacial and Sub-Aerial Erosion and Sedimentation During Rapid Deglaciation
A vibrant dimension of the coupled climate-tectonics-erosion system is the potential impact of climate change on erosion rates due to differences in efficiency of glacial and non-glacial erosion processes. In this context, glacial erosion rates inferred from sediment budgets in the recently deglaciated fjords of SE Alaska have received much attention as they are among the highest reported erosion rates worldwide. To improve understanding of these rates, and of the sediment records in these fjords more quantitatively, we developed a numerical model designed to examine the influence of terminus retreat on sediment yields from tidewater glaciers. We applied our model to Taan Fjord in Icy Bay, SE Alaska, where the Tyndall Glacier has retreated 17 km since 1959. Seismic surveys of the post glacial sediments in Taan Fjord demonstrates that 5.62 x 10$^{8}$ m$^{3}$of sediment was deposited in Taan Fjord between 1962 to 1999; the flux of sediment into the fjord averaged 1.52 x 10$^{7}$ m$^{3}$/a. If we assume that this sediment was entirely produced by Tyndall Glacier, this flux would correspond to a basin-averaged erosion rate of 45 mm/a for the past 40 years. These rates are too high, however, in part because a significant portion of the sediments are derived not from the glacier, but from prominent non-glaciated tributary streams evacuating sediment that was stored adjacent to the glacier during fjord occupation (Meigs et al., 2002). The coarser sediments are recognizable as alluvial fans and deltas that are distinct visibly and in the seismic profiles, but the finer sediments that travel further in the fjord tend to be undistinguishable seismically from distal glaciomarine deposits. To determine the volume of this finer sediment filling the fjord, we compared the volume of sediment evacuated from the dominant tributary valley adjacent to the current glacier terminus, to the volume contained in both the subaerial fan and submarine delta deposit. The total volume removed from the valley is 1.57 x 10$^{8}$ m$^{3}$, with 4.51 x 10$^{7}$ m$^{3}$ and 4.15 x 10$^{7}$ m$^{3}$ deposited in the fan and delta, respectively, and by inference, 7.04 x 10$^{7}$ m$^{3}$ deposited distally. Assuming the same ratio of proximal and distal deposition of sediment from a second major tributary, we estimate that an additional 1.71 x 10$^{7}$ m$^{3}$ of subaerial sediment was deposited distally in the fjord. Accounting for these subaerial contributions, which amount to 15% of the total sediment volume in Taan Fjord, the revised basin-averaged erosion rate for Tyndall Glacier for the past 40 years is 38 mm/a. Our model of proglacial sedimentation reveals a strong correlation between rapid glacial retreat and high glacial sediment yields, reflecting the strong tendency for ice fluxes to increase with retreat rates. This implies that most sediment yield data from tidewater glaciers in Alaska correspond to erosion rates that are higher than in the long term because rapid retreat has been characteristic of the period of study. Any modeling of landscape evolution in SE Alaska must thus take into account the impact of glacial advance-retreat cycles on long-term regional erosion. Extrapolating our correlation to a retreat rate of zero yields an estimate of long term glacial erosion rates between 7 and 13 mm/a, a factor of 4 +/- 1 less than current rates, and approaching modeled rates of regional tectonic uplift which reach 7 mm/a (Bird, 1996).
T31B-1297 0800h
Sediment flux history of Pearl River mouth basin, North margin of South China Sea
This work estimates the solid sediment flux in Pearl River mouth basin from Cenozoic (42Ma). The estimates were derived from isopach maps, seismic reflection profiles and drill holes. Average solid sediment fluxes were calculated for six epochs approximately corresponding to geological periods: Eocene-Lower Oligocene (42-29.3), Upper Oligocene (29.3-23.8), Lower Miocene (23.8-16.4), Middle Miocene (16.4-11.2), Upper Miocene (11.2-5.32), and Pliocene-Pleistocene (5.32-0). The total sediment flux from 42 Ma is 392071.3 km3 and 0.89 km of erosion formed from the onshore drainage basin area. The average erosion rate is 22 m/ Ma. The sediment flux curve shows 3 episodes massive increase in sediment supply, i.e. Upper Oligocene, Middle Miocene and Pliocene-Pleistocene. The first increase related to the break up activity and is the product of elevated rift shoulder. The other two increase peak link to the changing of climate.
T31B-1298 0800h
Cenozoic tectonics of the easternmost Tibetan plateau: Constraints from fission\-track geochronology
The Tibetan Plateau and surrounding regions are the result of the India-Asia collision that started at ca. 55 Ma, doubled the crust beneath Tibet (current thickness 60-70 km), and elevated the topography to 4-5 km. While thickening and uplift of the Plateau in front of the intender is mainly attributed to the crustal shortening, it has been proposed that crustal thickening east of the eastern Himalayan syntaxis, i.e. in eastern Tibet, is mainly the result of lower crustal flow. A critical test for this model is the quantification of age and distribution of uplift in eastern Tibet. Seventy new apatite fission-track ages from the eastern Tibetan Plateau and the Qinling are presented. Varying Cretaceous to Miocene ages across small vertical distances occur at the higher elevations of the Tibetan Plateau and in the entire Qinling. They are the result of long-lasting tectonic quiescence and peneplanation that governed the area since at least the late Cretaceous. Significant Miocene to Recent exhumation occurred along the topographic front of the central eastern Plateau (Longmen Shan) and the interior of the Songpan-Garze fold belt west of the Longmen Shan. Differences in the fission-track ages and the amount of exhumation between blocks separated by west-northwest dipping thrusts suggest out-of-sequence thrusting within the Longmen Shan belt and the Min Shan uplift zone. For the Min Shan the onset of exhumation is estimated from the age vs. elevation diagram at 10 - 16 Ma, the other areas yielded lower limits (7 - 11 Ma in central eastern Tibet) or upper limits (13 - 30 Ma in south-eastern Tibet) for the onset of exhumation. In the Qinling fast exhumation is not recorded in the apatite fission-track ages but in the track-length distributions of samples from the south-western part, i.e. the triple junction between the North China Block, the South China Block, and the Songpan-Garze. This indicates that fast exhumation took place but did not exceed 2 - 3 km. If exhumation is driven by lower crustal flow, it seems that the flow front has reached the central eastern Plateau margin during the middle Miocene causing significant denudation of the entire Plateau. The area north of the Longmen Shan is little denudated but the track length distributions of samples from the south-eastern Qinling reflect divergence of flow around the Longmen Shan and active Plateau growth in this area.
T31B-1299 0800h
Late Cenozoic Evolution of Jishi Range in Northeastern Tibet
Jishi Range is located in the northeastern margin of Tibetan Plateau, and is the southeastern end of Laji Mountain. This Range divides the Xunhua basin and Linxia basin. Prior to Jishi range uplift, these two basins and the range were united and covered by about 1000m Cenozoic sediments. The Growth strata in Linxia basin and Xunhua basin suggest deformation happened at about 8Ma B.P. or earlier. The detrital apatite fission track age sampled from Linxia basin show a recycled pattern and suggest the recycle occurred at about (5-13)Ma B.P. Furthermore, apatite fission track modeling results suggest a rapid cooling at about same time. All these suggest that the Jishi range uplifted from about (5-13) Ma B.P. At the onset of this uplift and subsequently erosion, the sediment was quickly removed, causing sediment recycle and fast cooling, but relatively slow surface uplift. The surface uplift rate was enhanced when the resistant bed rock was exhumed to surface. At about (4-2)Ma, the frequent paleo-soil deposited in Xunhua basin. However, conglomerate deposited in Linxia basin since about the same time. These suggest the Jishi range uplifted high enough at about (4-2)Ma to establish the present semi-arid condition, and the relief precipitation could result in conglomerate deposition.
T31B-1300 0800h
Nd and O Isotope Evidence for Middle Miocene Exhumation and Climate Change Along the NE Margin of Tibet
Oligocene to Pliocene sedimentation in the Linxia basin has been interpreted to reflect deposition in a flexural basin associated with contractional deformation along the northeastern margin of the Tibetan plateau. We use Nd isotopic and trace element data to characterize source rocks of northeastern Tibet and to determine the provenance of sedimentary rocks in the Linxia basin. Paleozoic-Early Mesozoic metasedimentary source rocks from the Kunlun-Qaidam and Songpan-Ganzi terranes have \epsilon$_N$$_d$ = -11.8 to -16.1; Paleozoic and Mesozoic plutonic source rocks have \epsilon$_N$$_d$ = -3.6 to -11.2; and Cretaceous sedimentary source rocks have \epsilon$_N$$_d$ = -9.7 to -9.9. Wind-blown dust derived from Central Asian deserts, another possible source of fine-grained deposits, displays \epsilon$_N$$_d$ values = -8.6 to -10.5 in the Loess plateau and North Pacific (Gallet et al., 1996; Pettke et al., 2000). Sedimentary rocks of the Linxia basin are less negative than metasedimentary rocks within the margin of the Tibetan plateau and range between -8.4 and -10.4 before 15 Ma and -6.2 and -11.8 after 14 Ma. The relatively positive \epsilon$_N$$_d$ values of Linxia basin mudstones compared to metasedimentary source rocks suggests that loess has dominated the fine-grained sedimentary budget throughout the basin's history. The increase in the range of \epsilon$_N$$_d$ values by 14 Ma is consistent with additional contributions from metasedimentary and plutonic source rocks in the margin of the plateau and may reflect an increase in erosion of local sources. This inference is supported by several other lines of evidence that significant exhumation occurred at approximately the same time, including (U-Th)/He analysis of apatite from plutonic rocks in the hanging wall of the West Qinling fault (Clark et al., this session) and detrital fission track data from Linxia basin deposits (Zheng et al., 2003). Increased rates of exhumation in the margin of the Tibetan plateau between 14 and 12 Ma are in agreement with the timing of a significant change in climate between 13 and 12 Ma in Linxia basin. A positive shift of 1.5\permil in the most negative \delta$^1$$^8$O values of lacustrine carbonates in Linxia basin between 13 to 12 Ma may reflect reorganization of atmospheric circulation and an increase in aridity on the NE margin of the Tibetan plateau perhaps associated with the plateau achieving an elevation sufficient to block moisture from the Indian and/or Pacific Oceans. Similar timing between exhumation and climate change suggests northeastward propagation of the plateau margin was responsible for the middle Miocene climate change observed in the Linxia basin.
T31B-1301 0800h
Low Temperature Thermochronomtry in the Western Foothills Fold-and-Thrust Belt: The Importance of Structure
Understanding the development of fold-and-thrust belts is a key component of understanding the feedbacks between erosion and tectonics at the orogen scale. In Taiwan, the western foothills fold-and-thrust belt accommodates over half the total convergence between the continental margin and the Philippine Sea plate. The west-verging asymmetric fault-related folds and imbricate thrusts that deform the tertiary sediments of the western foothills are a classic example of a fold-and-thrust belt, and an ideal place to study erosion, exhumation, and uplift in this tectonic setting. To date, we have had poor constraints on the interaction between surface erosion and individual structures. However, the location of samples relative to individual faults can play a major control on the cooling history recorded by a low temperature thermochronometer such as apatite fission track or U-Th/He dating. We present the results of samples collected along two transects in the Western foothills and explore the importance of thrust faults and the stratigraphic position of samples in controlling cooling histories. We combine a forward structural model, with a 2-dimensional finite-element thermo-kinematic model to establish the cooling history, and by inference, an exhumation and deformation history across structures in the fold-and-thrust belt. Our results show the importance of this type of combined modeling approach when using low temperature thermochronometers to constrain exhumation.
T31B-1302 0800h
Exhumation of Basement-Cored Uplifts: Example of the Kyrgyz Range Quantified with Apatite Fission-Track Thermochronology
The Kyrgyz range, the northernmost portion of the Kyrgyzstan Tian Shan, provides an example of a reverse-fault bounded mountain range with topographic evidence for progressive lateral propagation of surface uplift and exhumation. The highest and most deeply dissected central portion of the range lies south of Bishkek, in the region of the Ala Archa river. To the east and west, topography and relief decrease and remnants of a Cretaceous regional erosion surface imply minimal amounts of bedrock exhumation. Deciphering the timing of exhumation of segments of the range outward from the center measures the lateral propogation rate of the range-bounding reverse fault and quantifies the time and erosion depth needed to transform a mountain range from a juvenile to a mature morphology. New apatite fission-track (AFT) data from three transects from the eastern half of the Kyrgyz Range, combined with published AFT data, demonstrate that the range has propagated over 110 km from the presently highest region towards the east over the last 8-11 Ma. Based on the thermal and topographic evolutionary history of the Kyrgyz range, we present a model for a time-varying exhumation rate driven by changes in erodability, precipitation, and glaciation. Easily eroded, recently deposited sedimentary rocks overlying resistant basement control early, rapid exhumation and slow surface uplift rates. As increasing amounts of resistant basement are exposed, exhumation rates decrease while surface uplift rates increase, thereby growing topography. As the range becomes high enough to cause ice accumulation and orographically enhanced precipitation, fluvial and glacial erosion become more powerful and exhumation rates once again increase. Balance between erosion and uplift occurs through establishment of mature drainage networks into the range that connect high-elevation glaciated areas to deeply incised rivers.
T31B-1303 0800h
Walled Sedimentary Basins of China: Perpetrators or Victims of Plateau Growth?
Western China and adjacent areas of central Asia are characterized by low relief, internally drained sedimentary basins that are divided by actively uplifting mountain ranges. The margins of these basins often show evidence for extensive contractional deformation, yet their interiors are surprisingly stable. Basins such as the Tarim and Junggar also exhibit long and apparently continuous histories of closed drainage in the same approximate location (over 250 my in the case of Junggar). In contrast to traditional foreland basins, these basins are not uniquely associated with a specific thrust belt, nor do they show evidence for underlying decollements. We therefore propose the new term "walled basin", in recognition of the essential role of peripheral orogenic walls in creating and maintaining closed drainage and impounding sediments. Walled basins in Asia currently are restricted to areas that receive less than 40 cm/yr precipitation, suggesting that aridity plays a role in preventing fluvial breach of the basin walls (cf., Sobel et al., 2003). Entrapment of sediment within the closed Qaidam basin in the northeast Tibetan plateau has been implicated as a potential mechanism of plateau growth, based on the observations that the basin retains mass within the orogen and creates level topography. However, we propose that the Qaidam instead represents a walled basin that has been elevated due to underplating of the plateau, and is fated to eventual destruction as deformation continues. Several lines of reasoning support this conclusion. First, DEM analysis shows that modern drainage divides for the Qaidam and other walled basins never rise more than 1-2 km above the basin floors, limiting the amount of possible topgraphic infill. Second, the Tarim and Junggar basins presently remain well below 2000 m and probably have never been higher, despite receiving large influxes of detritus from adjacent ranges. Third, the Qaidam basin, like the Tarim and Junggar basins, has an older history of nonmarine fill that dates back at least to the Jurassic, and therefore its existence predates the Himalayan orogeny. Fourth, mid-Tertiary and older fill of the Qaidam basin has already been deformed, indicating an ongoing history of structural shortening. Finally, closed geomorphic basins within the southern Tibetan plateau are all much smaller than the Qaidam. This suggests that brittle deformation associated with progressive south to north underplating has disrupted preexisting sedimentary basins that were originally more prominent than they are now.
T31B-1304 0800h
Ain't No Mountain High Enough: Reconstructing Paleoelevation of Eroded Orogens
Quantitative paleoelevation reconstructions are a key element in understanding the relationships among uplift, erosion, and the demise of orogens. We present a new approach of determining paleoelevation based on the stable isotope composition of meteoric water that interacted with detachment mylonite during exhumation and uplift. Provided that this interaction is precisely dated and temporally linked to the stable isotope record in syntectonic basins this approach relates the tectonic, sedimentary and climate history of the orogen. Hydrogen, oxygen and Ar-Ar data from extensional detachments of the Shuswap, Kettle and Raft River core complexes and their adjacent basins, allow us to reconstruct discrete episodes in the Cenozoic elevation history of the North American Cordillera. Our data are consistent with a southward migration of a topographic high from Eocene to Miocene times, as reflected in the sequential onset of detachment faulting and extensional collapse in the Cordilleran hinterland. The integrated oxygen and hydrogen isotope data show that meteoric waters infiltrating the early to middle Eocene (49.0 - 47.0 Ma) eastern detachments of the Shuswap (British Columbia) and Kettle (Washington) core complexes had very negative hydrogen isotope values (-135 and -120 permil, respectively). These results require mean elevations in excess of 4000 m immediately preceding the timing of extensional deformation. Preliminary hydrogen (meteoric fluid compositions of -100 to -110 permil) and oxygen isotope data from Oligocene and Miocene mylonites of the Raft River complex (Utah) indicate that at lower latitudes high elevations persisted until Oligocene/Miocene times, a conclusion consistent with stable isotope data derived from syntectonic sedimentary deposits in the adjacent Elko Basin (Nevada). The coupled Eocene to Miocene data sets derived from rocks that at the time were at several km depth indicate that crustal thickening prior to detachment faulting resulted in high mean elevations which in turn controlled the local precipitation pattern and ultimately the local climate. It is conceivable that the Cenozoic North American Cordillera resembled the present-day Andean Puna-Altiplano plateau. Our results suggest that high mean elevations in the North American Cordillera existed through the Cenozoic with an Eocene to Miocene southward migration of orogenic collapse.
T31B-1305 0800h
Normal Faulting at the Western Margin of the Altiplano Plateau, Southern Peru
Although the western margin of the Altiplano Plateau is commonly used to illustrate the marked differences in the evolution of a mountain range with strong latitudinal and longitudinal precipitation gradients, the nature of tectonism in this semi-arid region is poorly understood and much debated. The western margin of the Altiplano in southern Peru and northern Chile marks an abrupt transition from the forearc region of the Andes to the high topography of the Cordillera Occidental. This transition has been interpreted by most workers as a monocline, with modifications due to thrust faulting, normal faulting, and gravity slides. Based on recent fieldwork and satellite image analysis, we suggest that, at least in the semi-arid climate of southern Peru, this transition has been the locus of significant high-angle normal faulting related to the block uplift of the Cordillera Occidental. We have focused our initial work in the vicinity of 15\deg S latitude, 71\deg W longitude, where the range front crosses Colca Canyon, a major antecedent drainage northwest of Arequipa. In that area, Oligocene to Miocene sediments of the Moquegua Formation, which were eroded from uplifted terrain to the northeast, presently dip to the northeast at angles between 2 and 10°. Field observations of a normal fault contact between the Moquegua sedimentary rocks and Jurassic basement rocks, as well as 15-m resolution 3-D images generated from ASTER satellite imagery, show that the Moquegua units are down-dropped to the west across a steeply SW-dipping normal fault of regional significance. Morphology of the range front throughout southern Peru suggests that normal faulting along the range front has characterized the recent tectonic history of the region. We present geochronological data to constrain the timing of movement both directly from the fault zone as well as indirectly from canyon incision that likely responded to fault movement.
T31B-1306 0800h
Detrital Chromian Spinel and Garnet From Bangladesh: Constraints on the Uplift History of the Himalayas
Bangladesh is in the northeastern part of the Bengal fan, which is composed of the detritus from the Himalayas. The Cenozoic sediments of the Bengal fan have recorded the uplift history of the Himalayas. This paper treats chemistries of detrital chromian spinels and garnets of the Eocene to Pleistocene sandstones in the Sylhet and Chittagong areas, which are located in northeastern and southeastern Bangladesh, respectively. We studied chemistries of detrital chromian spinels and garnets by EPMA to clarify the host rocks. The detrital chromian spinels were obtained from all the sandstones. Most of the detrital chromian spinels from the Oligocene to Pleistocene sandstones are very similar to those of serpentinites in the Yarlung-Zangbo ophiolite which was a suture zone between the Lhasa Block and Transhimalaya. However, the chemistry of detrital chromian spinels from the Eocene sandstone is somewhat different from that of the Yarlung-Zangbo ophiolite, and some of the detrital chromian spinels have similar chemistry with that of the spinels in the flood basalt in the Deccan Traps. This suggests that the Yarlung-Zangbo ophiolite has supplied detrital chromian spinels since the Oligocene. Garzanti et al. (1987) reported the occurrence of detrital chromian spinels from the lower Eocene in the western part of the Himalayas. Thus there was a time lag in the exhumation of ophiolitic rocks between the western and eastern parts of the Himalayas. The detrital garnets were obtained from most of the sandstones in the Sylhet and Chittagong areas. There are different tendencies of detrital garnet chemistries among sandstones from the Eocene to Oligocene, Early Miocene, and Middle Miocene to Pleistocene. As conclusions, it is inferred that the Tertiary granitic rocks in the Higher Himalaya started to supply detrital garnet to the Bengal fan in the Early Miocene, and the erosion of metamorphic rocks in the Lesser Himalaya seems to have started in the Middle Miocene.
T31B-1307 0800h
Triple Whammy: Mid-Holocene Landslide Dam Yields Suspended Load-Bedload Ratio, Regional Erosion Rate, and Bedrock Incision Rate, Central Nepal Himalaya
About 5400 years ago, a large landslide formed a $>$400-m-tall dam in the upper Marsyandi River, central Nepal. The resulting lake and delta deposits stretched $>$5 km upstream and reached a thickness of 140 m. $^{14}$C dating of 6 wood fragments reveal that the entire process of sedimentation and incision occurred remarkably quickly ($\sim$450 yr) with lake in-filling taking 240 $\pm$ 140 yr and reincision of the sediments largely complete after another 200 $\pm$ 140 yr. Based on a DEM and remnants of lake-related sediments, reconstructed volumes of the lake and delta deposits indicate a suspended load-to-bedload ratio of 2:1, with the fine sands and mud of the lake deposits representing $\sim$0.08 km$^{3}$ and the boulders, cobbles, and course sands in the delta constituting $\sim$0.04 km$^{3}$. Given the upstream area of 1600 km$^{2}$ and a depositional period of 240 yr, this total volume of 0.12 km$^{3}$ indicates a regional erosion rate of 0.3 $\pm$ 0.2 mm/yr. This upstream catchment lies primarily above the South Tibetan Detachment fault and within the modestly deformed Tethyan rocks. The preservation of these Tethyan strata north of the Himalayan crest is consistent with their gentler hillslopes and a considerably less vigorous erosional regime in comparison to the southern flank of the Himalaya. At the downstream end of the landslide dam, the Marsyandi River incised a new channel, cutting a slot canyon through at least 70 m of Greater Himalayan gneiss. This requires a minimum bedrock incision rate of 13 mm/yr over last 5400 yr, though presumably the majority of the incision occurred over a small fraction of the time and led to short-term incision rates $\ge$2 times higher.
T31B-1308 0800h
Constraining the Tectonic History of the Himalaya in Central Nepal With Low-Temperature Thermochronometry and Numerical Modeling
Previous studies have suggested that all recent deformation in the Himalayas has taken place at the southernmost frontal thrust of the orogen. However, recent studies suggest Plio-Quaternary displacement near the Main Central Thrust (MCT) in Central Nepal. Resolving this conflict is essential in understanding the role of climate change in the Late Cenozoic evolution of the Himalayas. We obtained 67 apatite fission-track (AFT) cooling ages from a 40x40 km$^2$ area in the Marsyandi River catchment (e.g. Blythe et al., this volume). The range of sample elevations is 500-5,000 m including four vertical transects. Cooling ages range from 0-3.8 Ma, and ages vary by as much as 1.8 My at equal elevations within the study area. Cooling ages of exhumed samples are sensitive to their thermal history, which is a function of the pathway rocks take to the surface. We consider end-member models for sample exhumation of (1) pure erosion with no thrust faulting, and (2) erosion with simultaneous thrust motion. Pathways due to erosion are vertical, whereas pathways with active tectonics are parallel to fault motion. Because of this difference, the spatial distribution of cooling ages is diagnostic of the exhumation history, which can be simulated using numerical models to quantify the relative contributions of erosion and tectonics. We simulate tectonic and erosional histories of samples using a 3D thermo-kinematic finite-element model in order to determine whether or not the signal present in our data requires recent tectonic activity. Free parameters in the model include uplift timing, erosion rate, rate of fault motion, and fault dip angle. Exhumed sample thermal histories for each simulation are recorded in order to predict AFT cooling ages using a kinetic annealing algorithm. Models are run iteratively to explore the range of free parameters and determine the best-fit solutions with a statistical comparison. Age versus elevation plots for our erosional models show that predicted and observed ages match only for high ($>$5 mm/yr) erosion rates. Work in progress is evaluating the role of active thrust faulting on cooling ages.
T31B-1309 0800h
Formation of Exfoliation Joints
The Earth's internal stresses interact with the topographic surface to affect many phenomena. Exfoliation joints, or sheeting joints, are widespread manifestations of this interaction. These opening-mode fractures form subparallel to the Earth's surface, bounding roughly concentric slabs of rock that resemble the layers of an onion. They occur worldwide in all major bedrock types, attain in-plane dimensions of hundreds of meters, exert a strong influence on groundwater flow, and help produce spectacular scenery, as in Yosemite National Park. The mechanism that causes them has been enigmatic. They are widely regarded as forming in response to "removal of overburden", but large fractures do not open in rocks merely by relieving a compressive stress. High fluid pressures, thermal effects, rock heterogeneity, and weathering also are rejected as primary causes of these fractures. Tensile stresses normal to the surface are required for large exfoliation fractures to open. Intriguingly, high surface-parallel compressive stresses are widely documented where exfoliation joints occur. Both numerical and analytical solutions for two-dimensional elastic bodies show that localized tensile stresses perpendicular to the ground surface must develop beneath certain topographies subject to strong compressive stresses parallel to the surface. This highly non-intuitive effect reflects the profound influence that topography can have on stresses near the surface of the Earth, and it can explain how exfoliation joints open. The theoretical results also indicate that exfoliation joint distributions could be used to infer the horizontal stresses near the Earth's surface.
T31B-1310 INVITED 0800h
Thrust fault systems of the Nankai Trough accretionary wedge interpreted from 3-D seismic reflection images
In 1999 we acquired a large volume of 3-D seismic reflection data to image the deformational structure of the Nankai Trough accretionary wedge that forms due to the subduction of the Philippine Sea plate beneath Japan. We imaged an 8 x 80 km$^{2}$ area south of the Muroto Peninsula offshore Shikoku Island, southwestern Japan, from the trench across the seaward most 70 km of the accretionary wedge. These data have been processed with 3-D prestack time migration and in selected locations with 3-D prestack depth migration. The 3-D volume reveals complex thrust fault systems that have offset and displaced stratigraphic horizons at a range of scales from meters to kilometers. We identify four distinctive fault systems with increasing scale as: 1) small thrusts with lengths of 100 - 500 m and offsets of 10 - 100 m, many of these lie immediately adjacent and parallel to the primary accretionary wedge thrust system (\#2 below) or are conjugate thrusts (backthrusts), 2) the primary thrusts of the accretionary wedge imbricate thrust system, which initiate at the deformation front and develop into 1 - 5 km long thrusts with displacements of 100 - 1000 m, 3) thrusts (out-of-sequence) that develop after the initial imbricate thrust system develops (\#2 above), these are $>$ 5 km long, have a shallower dip than the imbricate thrust system and intersect the imbricate thrust system, and 4) the main plate-boundary thrust, or d\'{e}collement, that lies at the base of the accretionary wedge, which can be traced from its initiation at the trench to $>$ 50 km down into the subduction zone. Each of these thrust systems controls the accretionary wedge tectonics and influences the structure of the accretionary wedge in both the dip and strike directions. We will present evidence to show how these fault systems affect wedge tectonics. We will also show that surprisingly, the wedge thrust activity reacts to the regional stress pattern created by recent seamount subduction, centered 50 km northeast of the survey area, as much or more than the localized stress conditions produced by changes in the geometry of the d\'{e}collement or the top of the subducting oceanic crust.
T31B-1311 0800h
Tectonic and climatic control on river profiles for rivers draining northwards from the Pamir and Kunlun (Central Asia).
Collision orogens developed between two plates result not only in shortening, uplift and erosion of the rocks, but also compression, uplift and modification of the drainage systems.Many studies now relate orogenic uplifts to the interaction of plate compression with isostatic changes due to active denudation (England and Molnar, 1990). In this paper I outline the relationships between river profiles, drainage patterns, tectonics and climate during the indentation of Asia in the Pamir range and adjacent areas: it extends a previous study of rivers draining south (Brookfield, 1998). The reasons for choosing the Pamir and Kunlun are the following. a) The indentation is relatively simple and can thus be modelled with a relatively simple rigid indentation model. The major complication is due to the different behaviour of the western and eastern edges of the indenter. The western edge involves mostly ductile deformation of the Tadjik back-arc basin to form a fold and thrust belt. The eastern edge involves strong shearing between continental crust of the Pamir and Tarim basins to form a complex collisional transform zone (marked by the Karakoram and associated faults) linking the Pamir arc with the Kunlun and Himalaya. b) The compression pattern is relatively simple and various tectonic units can mostly be traced from west to east across the Pamir indenter. Individual tectonic elements and ancient sedimentary basins can be followed almost continuously from the hardly compressed Afghan area through the highly compressed Pamir indent into the less compressed Kunlun and Tibetan plateau area. c) The displacements are enormous, relatively recent, and measurable. The Pamir arc only started developing in the Miocene around 20 ma. Since then over 800 km of internal shortening has occurred between the Indian shield and the Tien Shan(Dewey et al., 1989). Most of this post-Oligocene shortening occurred in the Pamir arc itself. And because of this, the earlier progressive Paleocene - Oligocene collisions of India with magmatic arcs south of Asia can be followed in some detail in the Pakistan Himalaya though not in the Indian Himalaya. d) The river profiles and courses can be directly related to the major tectonic development of the arc, modified by the influence of Quaternary climatic change (Molnar and England, 1990). The main drainage divide is along the crest of the fundamentally Mesozoic Hindu Kush and Karakoram ranges and extensions. Despite the late Cenozoic uplift of the Pamir, only the Pyandzh river cuts across the Pamir range in a course that corresponds with a geophysical but not a geological boundary. The rest of the rivers, with a few exceptions, tend to run in valleys parallel to the arc, except to the west and east. To the west, in northern Afghanistan the rivers still run northward from the westward extension of the Hindu Kush. To the east the main rivers have headwaters far within the Tibetan plateau and cut, with incredibly steep gradients across the Kun Lun and related ranges - testifying to the latest Tertiary development of this range. REFERENCES Brookfield, 1998. The evolution of the great river systems of southern Asia during the Cenozoic India-Asia collision: rivers draining southwards. Geomorphology, 22: 285-312. Dewey, J.F., Cande, S. and Pitman III, W.C., 1989. Tectonic evolution of the India/Eurasia collision zone. Eclogae geologica Helvetica, 82: 717-734. England, P. and Molnar, P., 1990. Surface uplift, uplift of rocks and exhumation of rocks. Geology, 18: 1173-1177. Molnar, P. and England, P., 1990. Late Cenozoic uplift of mountain ranges and global climatic change: chicken or egg? Nature, 346: 29-34.
T31B-1312 0800h
Spatially Concentrated Erosion Focuses Deformation Within the Himalayan Orogenic Wedge: Sutlej Valley, NW Himalaya, India
Long-term erosion processes in the NW-Himalaya have not only shaped the distribution of topography and relief, but may also exert a regional control on the kinematic history of the Himalayan orogenic wedge. The topographic front of the orogenic wedge forms the southern margin of the High Himalaya and may be related to subsurface structures such as a crustal ramp or a blind thrust. Drastic along- and across-strike erosional gradients characterize the modern Himalaya and range from high-erosion regions along the southern High Himalayan front where monsoonal precipitation is able to penetrate far into the range, to low-erosion sectors across the moderately elevated Lesser Himalaya to the south and the high-elevation, arid sectors to the north. Published paleo-elevation estimates from the Thakkhola Graben (Nepal) suggest that by ~11 Ma the southern Tibetan Plateau and probably the High Himalaya had been uplifted to elevations comparable to the recent conditions. Thus, the presently observed pronounced erosional gradients have likely existed across the orogen since then. However, the cause of high rock-uplift and exhumation rates along distinct segments of the southern front of the High Himalaya are still a matter of debate. New apatite fission track (AFT) and $^{40}$Ar/$^{39}$Ar data sampled along an orogen-perpendicular transect following the Sutlej Valley, approximately perpendicular to the Himalayan orogen, constrain the distribution patterns of rapid cooling related to rock uplift and exhumation. Combined with published thermochronologic data, this comprehensive AFT dataset from south of the High Himalaya mountain front to the interior of the Tethyan Himalaya allows us to derive a regional uplift and exhumation scenario. Our new $^{40}$Ar/$^{39}$Ar ages ranging between 17 and 4 Ma reveal diachronous exhumation of two crystalline nappes (Higher and Lesser Himalayan crystalline) during Miocene-Pliocene time. In contrast, the AFT data ranging from 1.3 to 4.6 Ma indicate synchronous, fast rock uplift and exhumation of both units. AFT ages increase with increasing elevation and distance from the Sutlej. Importantly, the AFT ages are particularly young where monsoonal precipitation is concentrated and the deeply incised Sutlej River crosses the southern front of the High Himalaya, thus indicating focused exhumation. The gradual change in AFT ages across and along strike of the major tectonic boundaries (STDS, MCT) in the High Himalaya shows that apparently none of them has been reactivated in the Sutlej area during Pliocene - Quaternary time. Contrary to previous interpretations we suggest that differential uplift is accommodated by widely distributed and pervasive brittle deformation rather than large-scale structures. We therefore conclude that the locus of maximum exhumation coincides with a reorganization of the orogenic wedge in order to compensate the erosional loss of material in the sector with focused precipitation.