Search

Rivers transfer terrestrial organic carbon (OC) from mountains to ocean basins, playing a key role in the global carbon cycle. During fluvial transit, OC may be oxidized and emitted to the atmosphere as CO2 or preserved and transported to downstream depositional sinks. The balance between oxidation and preservation determines the amount of particulate OC (POC) that can be buried long term, but the factors regulating this balance are poorly constrained. Here, we quantify the effects of fluvial transit on POC fluxes along an ~1,300 km lowland channel with no tributaries. We show that sediment transit time and mineral protection regulate the magnitude and rate of POC oxidation, respectively. Using a simple turnover model, we estimate that annual POC oxidation is a small percentage of the POC delivered to the river. Modelling shows that lateral erosion into POC-rich floodplains can increase POC fluxes to downstream basins, thereby offsetting POC oxidation. Consequently, rivers with high channel mobility can enhance CO2 drawdown while management practices that stabilize river channels may reduce the potential for CO2 drawdown., Nature Geoscience, 14 (11), ISSN:1752-0908, ISSN:1752-0894

In situ cosmogenic nuclides are an important tool for quantifying landscape evolution and dating fossil-bearing deposits in the Cradle of Humankind (CoH), South Africa. This technique mainly employs cosmogenic 10-Beryllium (Be-10) in river sediments to estimate denudation rates and the ratio of 26-Aluminium (Al-26) to Be-10 (Al-26/Be-10), to constrain ages of sediment burial. Here, we use Be-10 and Al-26 concentrations in bedrock and soil above the Rising Star Cave (the discovery site of Homo naledi) to constrain the denudation rate and the exposure history of soil on the surface. Apparent Be-10-derived denudation rates obtained from pebble- to cobble-sized clasts and coarse-sand in soil (on average 3.59 +/- 0.27 m/Ma and 3.05 +/- 0.25 m/Ma, respectively) are 2-3 times lower than the bedrock denudation rates (on average 9.46 +/- 0.68 m/Ma). In addition, soil samples yield an average Al-26/Be-10 ratio (5.12 +/- 0.27) that is significantly lower than the surface production ratio of 6.75, which suggests complex exposure histories. These results are consistent with prolonged surface residence of up to 1.5 Ma in vertically mixed soils that are up to 3 m thick. We conclude that the Be-10 concentrations accumulated in soils during the long near-surface residence times can potentially cause underestimation of single-nuclide (Be-10) catchment-wide denudation rates in the CoH. Further, burial ages of cave sediment samples that consist of an amalgamation of sand-size quartz grains could be overestimated if a pre-burial Al-26/Be-10 ratio calculated from the surface production is assumed.

Quantifying the time scales of sediment transport and storage through river systems is fundamental for understanding weathering processes, biogeochemical cycling, and improving watershed management, but measuring sediment transit time is challenging. Here we provide the first systematic test of measuring cosmogenic meteoric Beryllium-10 (10Bem) in the sediment load of a large alluvial river to quantify sediment transit times. We take advantage of a natural experiment in the Rio Bermejo, a lowland alluvial river traversing the east Andean foreland basin in northern Argentina. This river has no tributaries along its trunk channel for nearly 1,300 km downstream from the mountain front. We sampled suspended sediment depth profiles along the channel and measured the concentrations of 10Bem in the chemically extracted grain coatings. We calculated depth-integrated 10Bem concentrations using sediment flux data and found that 10Bem concentrations increase 230% from upstream to downstream, indicating a mean total sediment transit time of 8.4 ± 2.2 kyr. Bulk sediment budget-based estimates of channel belt and fan storage times suggest that the 10Bem tracer records mixing of old and young sediment reservoirs. On a reach scale, 10Bem transit times are shorter where the channel is braided and superelevated above the floodplain, and longer where the channel is incised and meandering, suggesting that transit time is controlled by channel morphodynamics. This is the first systematic application of 10Bem as a sediment transit time tracer and highlights the method's potential for inferring sediment routing and storage dynamics in large river systems., Postprints der Universität Potsdam : Mathematisch-Naturwissenschaftliche Reihe; 1119

The cosmogenic meteoric 10Be that is produced in the atmosphere and mainly scavenged by rainfall is a valuable tool for determining dates and rates of Earth surface processes. A key prerequisite for its applications is the knowledge of the long-term 10Be depositional flux to Earth's surface. Previous efforts on obtaining 10Be depositional fluxes include general atmospheric circulation modelling (GCM), empirical fitting of rainfall 10Be records, and determining the inventory of soil profiles of known age. Here, we derive 10Be depositional fluxes from Holocene river terrace 10Be inventories across the Taiwan mountain belt. We measured 10Be and 9Be concentrations from three terrace profiles, with 14C ages ranging between 3.83 and 9.05 cal. kyr BP and sampling depths of up to 6.2 m. Based on 10Be inventories of our river terraces, the calculated long-term 10Be depositional fluxes vary from 0.32 to 0.49 × 106 at/cm2/yr in the Taiwan orogen. We identify partial Be retention, surface erosion, multi-phase sediment deposition, and incomplete sampling depth in these terraces as potential causes of flux underestimation. When accounting for each of these factors and their uncertainties, the resulting flux estimates of each terrace show an overlapping range of 0.66–0.88 × 106 at/cm2/yr. This newly constrained flux range representative over millennial timescales is thus recommended for Earth surface applications in the fast-eroding Taiwan orogen, especially in regions with similar climatic conditions as the studied terraces. When comparing to flux estimates from other approaches, the flux from the rainfall-based fitting equation exceeds our new terrace-derived flux by more than a factor of two. We consider such high flux to be unlikely for the Taiwan orogen, and suggest that the control of precipitation on the 10Be flux here may be overestimated in rainfall data fitting. The GCM-derived flux overestimates the 10Be depositional flux in this setting to a smaller degree (at least 23%), which is in accordance with findings from previous global data compilation of 10Be fluxes.

Advances in cosmogenic nuclide exposure dating have made moraines valuable terrestrial recorders of palaeoclimate. A growing number of moraine chronologies reported from the Central Andes show that tropical glaciers responded sensitively to past changes in precipitation and temperature over timescales ranging from 103 to 105 years. However, the causes of past glaciation in the Central Andes remain uncertain. Explanations have invoked insolation-modulated variability in the strength of the South American Summer Monsoon, teleconnections with the North Atlantic Ocean, and/or cooling in the Southern Hemisphere. The driver for these past climate changes is difficult to identify, partly due to a lack of dated moraine records, especially in climatically sensitive areas of the southern Central Andes. Moreover, new constraints are needed on precisely where and when glaciers advanced. We use cosmogenic 10Be produced in situ to determine exposure ages for three generations of moraines at the Sierra de Aconquija, situated at 27°S on the eastern flank of the southern Central Andes. These moraines record glacier advances at approximately 22 ka and 40 ka, coincident with summer insolation maxima in the sub-tropics of the Southern Hemisphere, as well as at 12.5 ka and 13.5 ka during the Younger Dryas and the Antarctic Cold Reversal, respectively. We also identify minor glaciation during Bond Event 5, also known as the 8.2 ka event. These moraines register past climate changes with high fidelity, and currently constitute the southernmost dated record of glaciation on the eastern flank of the Central Andes. To contextualise these results, we compile 10Be data reported from 144 moraines in the eastern Central Andes that represent past glacier advances. We re-calculate exposure ages from these data using an updated reference production rate, and we re-interpret the moraine ages by taking the oldest clustered boulder age (after the exclusion of outliers attributed to nuclide inheritance) as closest to the timing of glacier advance—an approach for which we provide empirical justification. This compilation reveals that Central Andean glaciers have responded to changes in temperature and precipitation. We identify cross-latitude advances in phase with insolation cycles, the last global glacial maximum, and episodes of strengthened monsoonal moisture transport including the Younger Dryas and Heinrich Stadials 1 and 2. Our results from the Sierra de Aconquija allow us to constrain the southerly limit of enhanced precipitation associated with Heinrich Stadials at ∼25°S. More broadly, our findings demonstrate at both local and regional scales that moraines record past climate variability with a fine spatial and temporal resolution.

The lower Amazon basin contains vast floodplains that exchange sediment with the main river. The exchange of sediment between the floodplain and the channel follows a seasonal cycle that is anticorrelated with the hydrological cycle. At low water stages, sediment that has been stored in the floodplain for potentially several thousands of years is eroded and transferred from the floodplain to the mainstem. During high water stages, most sediment transported in the main channel stems directly from the eroding source with little intermittent storage. We apply the meteoric cosmogenic 10Be to stable 9Be ratio (10Be/9Be) as a denudation and weathering proxy to investigate this seasonality in sediment transport. Single meteoric 10Be concentrations ([10Be]) have previously been shown to record floodplain storage; whereas fractions of mobilized 9Be, a trace metal released during weathering, provide degrees of weathering. The resulting 10Be/9Be ratio provides denudation rates of the sediment sources. We compare 10Be/9Be measured on suspended sediment from river depth profiles at two sites (Obidos/Macapa) and during two hydrological seasons (high/low water stage in 2013). We show that careful construction of river depth profiles in combination with Rouse modeling is necessary to exclude the grain size dependence that together with hydrodynamic sorting introduces a bias into single [10Be] and [9Be]. In contrast, the 10Be/9Be is largely uniform across grain sizes and is thus unaffected by sorting. Our 10Be/9Be ratios, the denudation rates derived from them, and mobilized 9Be fractions do not show variations outside their uncertainties across seasons or sites. Thus, 10Be/9Be show that upstream sediment sources were well mixed by multiple cycles of storage and remobilization in the floodplain.

Mountain rivers respond to strong earthquakes by rapidly aggrading to accommodate excess sediment delivered by co-seismic landslides. Detailed sediment budgets indicate that rivers need several years to decades to recover from seismic disturbances, depending on how recovery is defined. We examine three principal proxies of river recovery after earthquake-induced sediment pulses around Pokhara, Nepal's second largest city. Freshly exhumed cohorts of floodplain trees in growth position indicate rapid and pulsed sedimentation that formed a fan covering 150 km2 in a Lesser Himalayan basin with tens of metres of debris between the 11th and 15th centuries AD. Radiocarbon dates of buried trees are consistent with those of nearby valley deposits linked to major medieval earthquakes, such that we can estimate average rates of re-incision since. We combine high-resolution digital elevation data, geodetic field surveys, aerial photos, and dated tree trunks to reconstruct geomorphic marker surfaces. The volumes of sediment relative to these surfaces require average net sediment yields of up to 4200 t km–2 yr–1 for the 650 years since the last inferred earthquake-triggered sediment pulse. The lithological composition of channel bedload differs from that of local bedrock, confirming that rivers are still mostly evacuating medieval valley fills, locally incising at rates of up to 0.2 m yr–1. Pronounced knickpoints and epigenetic gorges at tributary junctions further illustrate the protracted fluvial response; only the distal portions of the earthquake-derived sediment wedges have been cut to near their base. Our results challenge the notion that mountain rivers recover speedily from earthquakes within years to decades. The valley fills around Pokhara show that even highly erosive Himalayan rivers may need more than several centuries to adjust to catastrophic perturbations. Our results motivate some rethinking of post-seismic hazard appraisals and infrastructural planning in active mountain regions.

The uplift of the Shillong Plateau, in northeast India between the Bengal floodplain and the Himalaya Mountains, has had a significant impact on regional precipitation patterns, strain partitioning, and the path of the Brahmaputra River. Today, the plateau receives the highest measured yearly rainfall in the world and is tectonically active, having hosted one of the strongest intra-plate earthquakes ever recorded. Despite the unique tectonic and climatic setting of this prominent landscape feature, its exhumation and surface uplift history are poorly constrained. We collected 14 detrital river sand and 3 bedrock samples from the southern margin of the Shillong Plateau to measure erosion rates using the terrestrial cosmogenic nuclide 10Be. The calculated bedrock erosion rates range from 2.0 to 5.6 m My−1, whereas catchment average erosion rates from detrital river sands range from 48 to 214 m My−1. These rates are surprisingly low in the context of steep, tectonically active slopes and extreme rainfall. Moreover, the highest among these rates, which occur on the low-relief plateau surface, appear to have been affected by anthropogenic land-use change. To determine the onset of surface uplift, we coupled the catchment averaged erosion rates with topographic analyses of the plateau's southern margin. We interpolated an inclined, pre-incision surface from minimally eroded remnants along the valley interfluves and calculated the eroded volume of the valleys carved beneath the surface. The missing volume was then divided by the volume flux derived from the erosion rates to obtain the onset of uplift. The results of this calculation, ranging from 3.0 to 5.0 Ma for individual valleys, are in agreement with several lines of stratigraphic evidence from the Brahmaputra and Bengal basin that constrain the onset of topographic uplift, specifically the onset of flexural loading and the transgression from deltaic to marine deposition. Ultimately, our data corroborate the hypothesis that surface uplift was decoupled from the onset of rapid exhumation, which occurred several millions of years earlier.

The Labrador Sector of the Laurentide Ice Sheet is characterized by a complex network of ice divides and an extensive landform record outlining two broad and opposing ice flows that are separated by a narrow Horseshoe Intersection Zone (HIZ). This geomorphic system gave rise to contrasting reconstructions, which reflect uncertainties on the temporal evolution of ice divides, the nature and age of the main landform systems and the overall pattern of ice retreat during the last deglaciation. Here, we address these issues through systematic mapping of glacial landforms and ice-flow indicators in a large area of northeastern Quebec and Labrador. The application of cosmogenic (10Be and 26Al) dating to esker-fed glaciomarine deltas and different rock surfaces brings new constraints on the chronological framework and insights on the subglacial thermal regime. Our results outline four main landform assemblages, two of which relating to a major system of opposite ice flows that delineate the position of the eastern Ancestral Labrador ice divide. Our reconstruction shows that the HIZ occurs several tens of km to the west of this divide and that these two features are genetically distinct. The HIZ relates to the late-glacial development of ice streams showing a massive convergent ice flow into Ungava Bay. The upstream extent of this ice flow system shows a strong coupling with the outline of Ungava Bay drainage divide, suggesting that the configuration of the HIZ could be the expression of a topographic control on the late-glacial ice sheet dynamics. Landforms also indicate a significant shift in the deglaciation mode, which evolved from an overall warm-based to an areally-confined cold-based ice retreat near the center of the former ice mass – a change in dynamics that also played a role in the configuration of the HIZ. Overall, areas indicative of cold-based ice conditions are limited to highly elevated terrains, thus ruling out models arguing for an extensive cold-based ice cover over Ungava Bay throughout the last deglaciation.

Terrestrial cosmogenic nuclide concentrations in fluvial deposits allow estimation of paleo-erosion rates and reconstruction of the response of landscapes to climatic perturbations. In partly ice-covered landscapes, however, incorporation of subglacially-derived sediments that were shielded by ice from cosmic can lead to erroneous erosion rate calculations. Here, we combine in situ-produced 10Be-derived erosion rates, based on sand and pebbles from a fluvial fill terrace and the modern riverbed in the upper Yamuna catchment, with numerical ice flow modelling to quantify this bias. New luminescence and surface exposure ages suggest that aggradation of the exposed deposits occurred between 29.9 ± 2.5 ka and 14.8 ± 2.8 ka. During most of the deposition, glaciers probably covered ∼19% of the catchment. 10Be concentrations of terrace sand samples differ from those of pebble samples. We obtained the lowest erosion rates from quartzite pebbles, which stem from low elevations, and the highest erosion rates from crystalline pebbles, which stem from high elevations in the Yamuna catchment. We explain these different erosion rates by differences in the steepness of the source areas, an effect that prevails throughout the entire aggradation period despite significant former ice-cover. Sand samples, which are thought to be derived from all elevation parts of the catchment, however show lower 10Be concentrations during the aggradation compared to present-day. We argue that this difference is due to a substantial subglacial origin of the sand during the aggradation period, and not necessarily related to enhanced erosion. We conclude that aggradation of the valley fill in the Yamuna catchment is most likely due to reduced discharge, and only marginally related to higher erosion rates during the late Pleistocene.

Recent developments in terrestrial cosmogenic nuclide (TCN) exposure dating and the reinterpretation of TCN boulder ages from moraines have improved our understanding of the glacial chronology in the Central Andes. According to these records, glacial advances throughout the region correlate with insolation-driven changes in the intensity of the South American Summer Monsoon and millennial-scale climate events such as Heinrich Stadials and the Younger Dryas. Quantifying the temperature and precipitation shifts during these events helps to constrain past moisture pathways and associated changes in atmospheric circulation patterns. Yet, particularly in the southern Central Andes, where a wealth of glacigenic landforms attests to formerly cooler and/or wetter conditions, the magnitudes of past temperature and precipitation changes are only loosely constrained at a few sites. Here, we present results from TCN-dated moraines combined with the reconstruction of former glaciers and paleolakes within the eastern sector of the southern Central Andes (24°–27°S) for the Last Glacial Maximum (LGM), Heinrich Stadial 1 (HS1), and the Younger Dryas (YD). We performed Monte-Carlo simulations with 2-dimensional models of ice flow and lake hydrology that are forced by a spatially-distributed surface energy balance model. Our results indicate that temperatures were 3.0–4.2 °C, 2.0–3.3 °C and 1.3–2.5 °C cooler during the LGM, HS1 and YD, respectively, in agreement with previous estimates elsewhere in the Central Andes. We find that temperature changes during the late glacial are compatible with sea-surface temperature anomalies derived from the tropical Atlantic. Precipitation was only 5–27% greater than today, which contrasts with larger anomalies reconstructed for the Bolivian part of the Andean Plateau (Altiplano). We attribute this discrepancy to the southerly position of our study region with respect to the Bolivian High, supporting the hypothesis that this atmospheric pressure system played a prominent role for South American Summer Monsoon dynamics during glacial episodes.

We analyze the source-to-sink variations of in situ 10Be, 26Al and 21Ne concentrations in modern sediment of the Po river catchment, from Alpine, Apennine, floodplain, and delta samples, in order to investigate how the cosmogenic record of orogenic erosion is transmitted across a fast-subsiding foreland basin. The in situ 10Be concentrations in the analyzed samples range from ∼ 0.8 × 10 4 at / g QTZ to ∼ 6.5 × 10 4 at / g QTZ . The 10Be-derived denudation rates range from 0.1 to 1.5 mm/yr in the Alpine source areas and from 0.3 to 0.5 mm/yr in the Apenninic source areas. The highest 10Be-derived denudation rates are found in the western Central Alps (1.5 mm/yr). From these data, we constrain a sediment flux leaving the Alpine and the Apenninic source areas (>27 Mt/yr and ca. 5 Mt/yr, respectively) that is notably higher than the estimates of sediment export provided by gauging (∼10 Mt/yr at the Po delta). We observe a high variability in 10Be concentrations and 10Be-derived denudation rates in the source areas. In the Po Plain, little variability is observed, and at the same time, the area-weighed 10Be concentration of ( 2.29 ± 1.57 ) × 10 4 at / g QTZ (±1 SD of the dataset) from both the Alps and the Apennines is poorly modified (by tributary input) in sediment of the Po Plain ( ( 2.68 ± 0.78 , ± 1 SD ) × 10 4 at / g QTZ ). The buffering effect of the Po floodplain largely removes scatter in 10Be signals. We test for several potential perturbations of the cosmogenic nuclide record during source to sink transfer in the Po basin. We find that sediment trapping in deep glacial lakes or behind dams does not significantly change the 10Be-mountain record. For example, similar 10Be concentrations are measured upstream and downstream of the postglacial Lake Maggiore, suggesting that denudation rates prior to lake formation were similar to today's. On the scale of the entire basin, the 10Be concentration of basins with major dams is similar to those without major dams. A potential modification of the 10Be-mountain record during sediment storage in the subsiding Po Plain can be excluded as measured 26Al/10Be ratios do not show the addition of deeply buried material. A barely resolvable excess 21Ne signal in the Po plain indicates recycling of previously eroded sediment rather than accumulation of cosmogenic nuclides during surficial floodplain transport. Our results demonstrate that the cosmogenic record of mountain erosion is effectively transmitted from the source areas to the sediment sink, even across a strongly subsiding foreland basin. We show that this record is poorly influenced by a range of potential geological and anthropogenic sources of bias, and is largely independent from upstream sediment interception and sediment storage in the floodplain.

Cosmogenic nuclide analysis in sediment from the Earth's largest rivers yields mean denudation rates of the sediment-producing areas that average out the local variations commonly found in small rivers. Using this approach, we measured in situ cosmogenic 26Al and 10Be in sand of >50 large rivers over a range of climatic and tectonic regimes covering 32% of the Earth's terrestrial surface. In 35% of the analyzed rivers, we find 26Al/10Be ratios significantly lower than these nuclides´ surface-production-rate ratio of 6.75 in quartz, indicating radioactive decay over periods exceeding 0.5 Myr. We invoke a combination of slow erosion, shielding in the source area, and sediment storage and burial during long-distance transport to explain these low ratios. In the other 65% of studied rivers we find 26Al/10Be ratios within uncertainty of their surface production-rate ratio, indicating cosmogenic steady state. For these rivers, we obtain a global source area denudation rate of 141 t/km2×yr (54 mm/kyr of rock-equivalent) that translates to a flux of 3.07 ± 0.56 Gt/yr. By assuming that this sub-dataset is representative of the global land surface, we upscale this value to the total surface area for exorheic basins, thereby obtaining a global denudation flux of 15.2 ± 2.8 Gt/yr that integrates over the past 11 kyr. This value is slightly lower than published values from cosmogenic nuclides from small river basins (23 (+53/−16)) Gt/yr) upscaled using a global slope model, and also lower than modern sediment and dissolved loads exported to the oceans (24.0 Gt/yr). Our new approach confirms an estimate of global dissolved and solid matter transfer that converges to an encouragingly narrow range of within 35%; whereas the use of paired nuclides in large rivers provides estimates of the buffering timescales of sediment transport.

We present an extensive investigation of a new erosion and weathering proxy derived from the 10Be(meteoric)/9Be(stable) ratio in the Amazon River basin. This new proxy combines a radioactive atmospheric flux tracer, meteoric cosmogenic 10Be, with 9Be, a trace metal released by weathering. Results show that meteoric 10Be concentrations ([10Be]) and 10Be/9Be ratios increase by >30% from the Andes to the lowlands. We can calculate floodplain transfer times of 2–30 kyr from this increase. Intriguingly however, the riverine exported flux of meteoric 10Be shows a deficit with respect to the atmospheric depositional 10Be flux. Most likely, the actual area from which the 10Be flux is being delivered into the mainstream is smaller than the basin-wide one. Despite this imbalance, denudation rates calculated from 10Be/9Be ratios from bed load, suspended sediment, and water samples from Amazon Rivers agree within a factor of 2 with published in situ 10Be denudation rates. Erosion rates calculated from meteoric [10Be], measured from depth-integrated suspended sediment samples, agree with denudation rates, suggesting that grain size-induced variations in [10Be] are minimized when using such sampling material instead of bed load. In addition, the agreement between erosion and denudation rates implies minor chemical weathering intensity in most Amazon tributaries. Indeed, the Be-specific weathering intensity, calculated from mobilized 9Be comprising reactive and dissolved fractions that are released during weathering, is constant at approximately 40% of the total denudation from the Andes across the lowlands to the Amazon mouth. Therefore, weathering in the Amazon floodplain is not detected.

Quantifying rates of weathering and erosion of mafic rocks is essential for estimating changes to the oceans alkalinity budget that plays a significant role in regulating atmospheric CO2 levels. In this study, we present catchment-wide rates of weathering, erosion, and denudation measured with cosmogenic nuclides in mafic and ultramafic rock. We use the ratio of the meteoric cosmogenic nuclide 10Be, deposited from the atmosphere onto the weathering zone, to stable 9Be, a trace metal released by silicate weathering. We tested this approach in stream sediment and water from three upland forested catchments in the north-west Czech Republic. The catchments are underlain by felsic (granite), mafic (amphibolite) and ultramafic (serpentinite) lithologies. Due to acid rain deposition in the 20th century, the waters in the granite catchment exhibit acidic pH, whereas waters in the mafic catchments exhibit neutral to alkaline pH values due to their acid buffering capability. The atmospheric depositional 10Be flux is estimated to be balanced with the streams’ dissolved and particulate meteoric 10Be export flux to within a factor of two. We suggest a correlation method to derive bedrock Be concentrations, required as an input parameter, which are highly heterogeneous in these small catchments. Derived Earth surface metrics comprise (1) Denudation rates calculated from the 10Be/9Be ratio of the “reactive” Be (meaning sorbed to mineral surfaces) range between 110 and 185 t km−2 y−1 (40 and 70 mm ky−1). These rates are similar to denudation rates we obtained from in situ-cosmogenic 10Be in quartz minerals present in the bedrock or in quartz veins in the felsic and the mafic catchment. (2) The degree of weathering, calculated from the fraction of 9Be released from primary minerals as a new proxy, is about 40–50% in the mafic catchments, and 10% in the granitic catchment. Lastly, (3) erosion rates were calculated from 10Be concentrations in river sediment and corrected for sorting and dissolved loss. These amount to 50% of denudation rates from 10Be/9Be in the mafic and ultramafic catchments, the remainder being mass loss in the dissolved form by weathering. In contrast, erosion comprises most of the mass loss in the granitic catchment. These first results are encouraging, given that we find overall good agreement between in situ and meteoric cosmogenic methods, that our denudation rates are in the range of those published for middle European river catchments, and that degrees of weathering are as expected for these diverse lithologies. This method allows quantifying rates of erosion and weathering in mafic rock over the time scale of weathering that are, unlike in situ cosmogenic 10Be, independent from the presence of quartz. 10Be/9Be therefore offers to quantify Earth surface processes in a wide range of landscapes underlain by mafic rock – rates that are of high importance for exploring climate-weathering feedbacks but that have been inaccessible to date.

Erosion and tectonic uplift are widely thought to be coupled through feedbacks involving orographic precipitation, relief development, and crustal weakening. In many orogenic systems, it can be difficult to distinguish whether true feedbacks exist, or whether observed features are a consequence of tectonic forcing. To help elucidate these interactions, we examine Gongga Shan, a 7556 m peak on the eastern margin of the Tibetan Plateau where cosmogenic 10Be basin-wide erosion rates reach >5 mm/yr, defining a region of localized rapid erosion associated with a restraining bend in the left-lateral Xianshuihe Fault. Erosion rates are consistent with topography, thermochronometry, and geodetic data, suggesting a stable pattern of uplift and exhumation over at least the past 2–3 My. Transpression along the Xianshuihe Fault, orographically enhanced precipitation, thermally weakened crust, and substantial local relief all developed independently in the Gongga region and existed there prior to the uplift of Gongga Shan. However, only where all of these conditions are present do the observed topographic and erosional extremes exist, and their relative timing indicates that these conditions are not a consequence of rapid uplift. We conclude that their collocation at 3–4 Ma set into motion a series of feedbacks between erosion and uplift that has resulted in the exceptionally high topography and rapid erosion rates observed today.

When using cosmogenic nuclides to determine exposure ages or denudation rates in rapidly evolving landscapes, challenges arise related to the small number of nuclides that have accumulated in surface materials. Improvements in accelerator mass spectrometry have enabled analysis of samples with low 10Be content (

Based on 10Be denudation rates previously found (3.6 m/Ma, [2]; 3.44 m/Ma, [1]), the landscape across the CoH is considered old and eroding slowly. High erosion rates similar to our results (5.13 - 15.02 m/Ma) for chert bedrock are ascribed to fast river incision or a recent partial collapse event [1,2]. In contrast, we think our high outcrop erosion rates reflect true denudation and low apparent values from soil samples indicate long retention of quartz on surface, while dolomite is largely removed in solution. The quartz then experiences periods of burial and reworking in caves and river terraces, resulting in low 26Al/10Be ratios.

Understanding past changes in the Antarctic ice sheets provides insight into how they might respond to future climate warming. During the Pliocene and Pleistocene, geological data show that the East Antarctic Ice Sheet responded to glacial and interglacial cycles by remaining relatively stable in its interior, but oscillating at its marine-based margin. It is currently not clear how outlet glaciers, which connect the ice sheet interior to its margin, responded to these orbitally-paced climate cycles. Here we report new ice surface constraints from Skelton Glacier, an outlet of the East Antarctic ice sheet, which drains into the Ross Ice Shelf. Our multiple-isotope (10Be and 26Al) cosmogenic nuclide data indicate that currently ice-free areas adjacent to the glacier underwent substantial periods of exposure and ice cover in the past. We use an exposure-burial model driven by orbitally-paced glacial–interglacial cycles to determine the probable ice surface history implied by our data. This analysis shows that: 1) the glacier surface has likely fluctuated since at least the Pliocene; 2) the ice surface was >200 m higher than today during glacial periods, and the glacier has been thicker than present for ∼75–90% of each glacial–interglacial cycle; and 3) ice cover at higher elevations possibly occurred for a relatively shorter time per Pliocene cycle than Pleistocene cycle. Our multiple-nuclide approach demonstrates the magnitude of ice surface fluctuations during the Pliocene and Pleistocene that are linked to marine-based ice margin variability.

The Song Gianh is a small-sized (~3500 km2), monsoon-dominated river in northern central Vietnam that can be used to understand how topography and climate control continental erosion. We present major element concentrations, together with Sr and Nd isotopic compositions, of siliciclastic bulk sediments to define sediment provenance and chemical weathering intensity. These data indicate preferential sediment generation in the steep, wetter upper reaches of the Song Gianh. In contrast, detrital zircon U-Pb ages argue for significant flux from the drier, northern Rao Tro tributary. We propose that this mismatch represents disequilibrium in basin erosion patterns driven by changing monsoon strength and the onset of agriculture across the region. Detrital apatite fission track and 10Be data from modern sediment support slowing of regional bedrock exhumation rates through the Cenozoic. If the Song Gianh is representative of coastal Vietnam then the coastal mountains may have produced around 132 000–158 000 km3 of the sediment now preserved in the Song Hong-Yinggehai Basin (17–21 of the total), the primary depocenter of the Red River. This flux does not negate the need for drainage capture in the Red River to explain the large Cenozoic sediment volumes in that basin but does partly account for the discrepancy between preserved and eroded sediment volumes. OSL ages from terraces cluster in the Early Holocene (7.4–8.5 ka), Pre-Industrial (550–320 year BP) and in the recent past (ca. 150 year BP). The older terraces reflect high sediment production driven by a strong monsoon, whereas the younger are the product of anthropogenic impact on the landscape caused by farming. Modern river sediment is consistently more weathered than terrace sediment consistent with reworking of old weathered soils by agricultural disruption.

Fluvial fill terraces in intermontane basins are valuable geomorphic archives that can record tectonically and/or climatically driven changes of the Earth-surface process system. However, often the preservation of fill terrace sequences is incomplete and/or they may form far away from their source areas, complicating the identification of causal links between forcing mechanisms and landscape response, especially over multi-millennial timescales. The intermontane Toro Basin in the southern Central Andes exhibits at least five generations of fluvial terraces that have been sculpted into several-hundred-meter-thick Quaternary valley-fill conglomerates. New surface-exposure dating using nine cosmogenic 10Be depth profiles reveals the successive abandonment of these terraces with a 100 kyr cyclicity between 75±7 and 487±34 ka. Depositional ages of the conglomerates, determined by four 26Al/10Be burial samples and U–Pb zircon ages of three intercalated volcanic ash beds, range from 18±141 to 936±170 ka, indicating that there were multiple cut-and-fill episodes. Although the initial onset of aggradation at ∼1 Ma and the overall net incision since ca. 500 ka can be linked to tectonic processes at the narrow basin outlet, the superimposed 100 kyr cycles of aggradation and incision are best explained by eccentricity-driven climate change. Within these cycles, the onset of river incision can be correlated with global cold periods and enhanced humid phases recorded in paleoclimate archives on the adjacent Bolivian Altiplano, whereas deposition occurred mainly during more arid phases on the Altiplano and global interglacial periods. We suggest that enhanced runoff during global cold phases – due to increased regional precipitation rates, reduced evapotranspiration, or both – resulted in an increased sediment-transport capacity in the Toro Basin, which outweighed any possible increases in upstream sediment supply and thus triggered incision. Compared with two nearby basins that record precessional (21-kyr) and long-eccentricity (400-kyr) forcing within sedimentary and geomorphic archives, the recorded cyclicity scales with the square of the drainage basin length. Fil: Tofelde, Stefanie. Universitat Potsdam; Alemania. Deutsches Geo Forschungs Zentrum; Alemania Fil: Schildgen, Taylor F.. Universitat Potsdam; Alemania. Deutsches Geo Forschungs Zentrum; Alemania Fil: Savi, Sara. Universitat Potsdam; Alemania Fil: Pingel, Heiko. Universitat Potsdam; Alemania Fil: Wickert, Andrew D.. University of Minnesota; Estados Unidos Fil: Bookhagen, Bodo. Universitat Potsdam; Alemania Fil: Wittmann, Hella. Deutsches Geo Forschungs Zentrum; Alemania Fil: Alonso, Ricardo Narciso. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tucumán. Instituto Superior de Correlación Geológica. Universidad Nacional de Tucumán. Facultad de Ciencias Naturales e Instituto Miguel Lillo. Departamento de Geología. Cátedra Geología Estructural. Instituto Superior de Correlación Geológica; Argentina Fil: Cottle, John. University of California; Estados Unidos Fil: Strecker, Manfred R.. Universitat Potsdam; Alemania

The ratio of the meteoric cosmogenic nuclide 10Be, precipitated from the atmosphere, to the stable nuclide 9Be, released by silicate weathering, was measured in suspended sediment of the Ganga River basin to determine denudation rates, degrees of weathering, and sediment storage in the floodplain. The 10Be precipitated and the 9Be released are carried to ca. 90% by amorphous and to 10% by crystalline Fe-hydroxides, as revealed by chemical extractions, whereas the dissolved Be transport is negligible due to the river pH of 8. Resulting 10Be/9Be ratios increase from 0.75 × 10 − 9 for the northern and Himalaya-draining rivers to 1.7 × 10 − 9 in the downstream basin. The increase in 10Be/9Be ratios results from two compounding effects: with 1) average denudation rates decrease from 0.5 mm yr−1 in the Himalayas to 0.17 mm yr−1 for the Ganga main stem in the lowlands, 2) the southern tributaries draining the low-relief craton contribute sediment with a ratio of 2.0 × 10 − 9 , corresponding to a denudation rate of 0.03 mm yr−1. We find that at the spatial scale of the entire basin, the atmospheric delivery flux of 10Be equals its sedimentary export flux. Hence fluxes can be considered to be at steady state and radioactive decay of 10Be during sediment storage is not discernible. The lack of a resolvable increase in 10Be concentration during sediment transfer along the floodplain stretch furthermore suggests that the sediment transfer time is indeed short. We also cannot resolve any additional release of silicate-bound 9Be there, testifying to the lower degree of weathering there. When multiplied with the basin area the 10Be/9Be-derived denudation rate of 0.14 mm yr−1 corresponds to a sediment flux of 350 Mt yr−1 which is in good agreement with gauging-derived sediment fluxes (ca. 400 Mt yr−1). Because they integrate over the entire basin, denudation rates from 10Be/9Be are lower than floodplain-corrected denudation rates from in situ cosmogenic 10Be that reflect the rates of the sediment-producing mountain areas.

The modern Amazon trunk stream and tributaries contain substantial amounts of sediments that were formerly stored for 1-3 m.y. in floodplains. By reworking of floodplain deposits, this sediment is being incorporated into Amazonian streams. We conclude this from using the ratio of in situ-produced cosmogenic Al-26 to Be-10 in detrital sediment from rivers draining the sediment-producing Andean, Guyana, and Brazilian shield areas, as well as the largest Amazonian lowland rivers. Samples from the modern Amazon River yield Al-26/Be-10 ratios between similar to 3.8 and 5.5, and thus represent mixing between a non-buried sediment end member and a formerly buried sediment end member. The non-buried end member is sourced in the Andes with a ratio of similar to 6.5, reflecting recent erosion of hillslopes exposed to cosmic rays, and the buried end member is potentially represented by deeply stored floodplain sediment as old as Miocene age, with a ratio of similar to 3, that was partially shielded from cosmic rays. This simple picture of binary mixing is complicated by grain-size effects. Coarser-grained sediment from non-Andean settings (lowlands, cratonic shields) contains slightly more formerly buried material, whereas finer-grained lowland sediment mostly records non-buried Al-26/Be-10 ratios and nuclide concentrations that are indicative of an Andean source. Our assessment shows an incorporation of formerly buried, old sediment of central Amazonian floodplains to the fresh, Andean-derived sediment mainly carried in the Amazon trunk stream. We interpret our results to mean that the Amazon regularly avulses into these old deposits that crop out in more distal vicinities to the main river and temporarily leaves its current Holocene channel.

The denudation rate signal of the Bolivian Andes as measured by cosmogenic 10 Be in sediment is preserved in the floodplain of adjacent foreland basins even though these basins store the sediment for thousands of years. This conclusion is drawn from comparing published Andean source area denudation rates with new cosmogenic 10 Be data as measured in the floodplains of the large Beni and Mamore basins. For the entire Beni basin including the sediment-producing Andes and the vast flooded plains of the foreland, the cosmogenic nuclide-derived denudation rate is 0.45 ± 0.06 mm/yr, while that of the Mamore basin is 0.55 ± 0.19 mm/yr. By comparison, the respective Andean source areas erode at averaged rates of 0.37 ± 0.06 mm/yr (upper Beni), and at 0.56 ± 0.09 mm/yr (upper Mamore). We notice a decrease in variability of denudation rate with increasing spatial scale as small-scale processes are averaged out. Sediment mixing within the floodplain damps scatter of nuclide-derived denudation rates picked up in the source area. On the temporal scale, a remarkable agreement between cosmogenic nuclide-derived rates averaging over a few kiloyears with those from fission track dating averaging over millions of years seems to suggest that cosmogenic nuclide-derived denudation rates capture the long-term erosional characteristics of the mountain belt. The sum of these observations suggests that any sample collected along a river traversing a floodplain will yield the denudation rate of the source area. This finding opens the unique possibility of constraining cosmogenic nuclide-derived paleo-sediment budgets for these large basins as the long-term, spatially-averaged denudation rate signal of the sediment-producing area is preserved in sedimentary archives.

The confidence in surface exposure dating and related research, such as erosion rate studies or burial dating, strongly depends on the accuracy and precision of the currently used production rates of in situ-produced cosmogenic nuclides. Reducing the uncertainties of nuclide production rates by more accurate calibrations with independently dated natural rock surfaces is crucial for further improving the quantification of earth surface processes. Here we use surface samples from the 760 ± 2 ka old Bishop Tuff in eastern California to quantify the 10 Be/ 21 Ne and 26 Al/ 21 Ne production rate ratios in quartz. Our determination is based on (1) measured nuclide concentrations of cosmogenic 10 Be, 21 Ne, and 26 Al, (2) a conservative estimate for the erosion of the tuff in the Volcanic Tableland area, which we base on our previously published 21 Ne concentrations [Goethals, M.M., Niedermann, S., Hetzel, R., Fenton, C.R., 2009. Determining the impact of faulting on the rate of erosion in a low-relief landscape: A case study using in situ produced 21 Ne on active normal faults in the Bishop Tuff, California. Geomorphology 103, 401–413] and a conservative estimate for the uncertainty of the 21 Ne production rate, and (3) the assumption of steady-state erosion. Other assumptions, such as the applied scaling procedure, the muon contribution to nuclide production, or the attenuation lengths of neutrons and muons in rock, do not substantially affect the results. Based on 13 samples, the following average production rate ratios and conservative uncertainty estimates are obtained for sea level, high latitude, open sky, and rock surface: 0.249 ± 0.009 or 0.232 ± 0.009 for 10 Be/ 21 Ne using 10 Be half-lives of 1.51 and 1.39 Ma, respectively, and 1.80 ± 0.09 for 26 Al/ 21 Ne (for an 26 Al half-life of 0.72 Ma). The 10 Be/ 21 Ne and the 26 Al/ 21 Ne production ratios are consistent with currently used production rates but the ratios are much more precise than previous determinations. The resulting 26 Al/ 10 Be production ratio of 7.23 ± 0.45 (for a 10 Be half-life of 1.51 Ma) or 7.76 ± 0.49 (for a 10 Be half-life of 1.39 Ma) is high when compared to previously published values. We discuss reasons for this difference, amongst them the possibility that 26 Al analyses in general might be compromised by artefacts affecting the stable Al concentration measurements. When combined with a 10 Be production rate of 5.01 at g − 1 a − 1 for the 1.51 Ma half-life (or 4.61 at g − 1 a − 1 for the 1.39 Ma half-life), our production ratios convert to 21 Ne and 26 Al production rates of 20.1 and 36.2 at g − 1 a − 1 (or 19.9 and 35.7 at g − 1 a − 1 ), respectively.

Measured from a bag of sand taken in large rivers, the concentration of cosmogenic nuclides such as in situ-produced 10Be, 26Al, and 14C can be used to constrain the mean sediment flux of the headwaters and assess the duration of sediment storage from source to sink. We revisit these principles, with examples from the Amazon and Ganga basins. We identify two end member cases controlling the concentration of cosmogenic nuclides in lowland river sediment: 1) if the time scale of floodplain sediment storage is short compared to the half-life of the nuclide, in situ cosmogenic nuclide concentrations are not significantly altered in lowland basins. In this case the concentration of e.g. in situ-produced 10Be in the sediment taken in the lowland basin equals in most cases that exported from the sediment source area, but the variability in nuclide concentrations between headwater streams is significantly averaged-out. Thus to convert the measured river sediments' in situ cosmogenic nuclide concentration into a catchment-wide denudation rate, production rates are scaled to include those of the sediment-producing mountainous areas only, rather than the entire catchment area. Nuclide production in the lowlands, where no sediment is being produced, is hence excluded. This correction is termed “floodplain correction”. 2) If sediment buried for periods of the order of the nuclides´ half-life is episodically re-entrained into the active river, paired nuclides, for example the ratio of 26Al/10Be, through the differential decay of these isotopes, constrain the the storage duration and re-mixing of floodplain sediment. As in-situ cosmogenic nuclides measure denudation rates over longer time scales, typically integrating between 103 and 105 years, their combination with modern estimates of sediment fluxes from river load gauging offers a rich potential of deciphering the controls of Earth surface fluxes across large basins. From this combination, we can assess how river systems react to external perturbations such as climate change or human interference, and how such signals are transmitted through the alluvial reaches of lowland basins. The most important results in both basins are that i) lowland Amazon and Ganga nuclide concentrations indeed broadly reflect the spatial average of Andean and Himalayan denudation, respectively, ii) the thus calculated sediment fluxes are within uncertainty of modern fluxes from sediment gauging, indicating complete sediment delivery to the sea without net loss into the basin, and iii) only the Amazon basin, given its size, contains a substantial fraction of sediment buried temporarily in the Quaternary. Finally, we provide a short introduction to the use of meteoric cosmogenic nuclide 10Be and its ratio to stable 9Be, released by rock weathering, in lowlands. In the Amazon basin we show that 10Be concentrations in fine-grained suspended loads, depth-integrated over the water column using Al/Si ratios, provide erosion rates whereas the 10Be/9Be ratio on Be adsorbed to particles provides the denudation rate. Both are similar to denudation rates from in situ 10Be. We show that this system responds more sensitively to sediment storage than the in situ system, and both accumulation and decay of meteoric concentrations may thus be used to determine sediment residence time.

Deciphering the response of sediment routing systems to climatic forcing is fundamental for understanding the impacts of climate change on landscape evolution. In the Kangra Basin (northwest Sub-Himalaya, India), upper Pleistocene to Holocene alluvial fills and fluvial terraces record periodic fluctuations of sediment supply and transport capacity on timescales of 103 to 105 yr. To evaluate the potential influence of climate change on these fluctuations, we compare the timing of aggradation and incision phases recorded within remnant alluvial fans and terraces with climate archives. New surface-exposure dating of six terrace levels with in-situ cosmogenic 10Be indicates the onset of incision phases. Two terrace surfaces from the highest level (T1) sculpted into the oldest preserved alluvial fan (AF1) date back to 53.4 ± 3.2 ka and 43.0 ± 2.7 ka (1σ). T2 surfaces sculpted into the remnants of AF1 have exposure ages of 18.6 ± 1.2 ka and 15.3 ± 0.9 ka , while terraces sculpted into the upper Pleistocene–Holocene fan (AF2) provide ages of 9.3 ± 0.4 ka (T3), 7.1 ± 0.4 ka (T4), 5.2 ± 0.4 ka (T5) and 3.6 ± 0.2 ka (T6). Together with previously published OSL ages yielding the timing of aggradation, we find a correlation between variations in sediment transport with oxygen-isotope records from regions affected by the Indian Summer Monsoon. During periods of increased monsoon intensity and post-Last Glacial Maximum glacial retreat, aggradation occurred in the Kangra Basin, likely due to high sediment flux, whereas periods of weakened monsoon intensity or lower sediment supply coincide with incision.

The Amundsen Sea Embayment of West Antarctica hosts one of the most rapidly changing sectors of the West Antarctic Ice Sheet. With the fastest-flowing ice streams in Antarctica, the region around Pine Island Bay is characterized by rapid ice-sheet thinning and grounding-line retreat. Published surface-exposure data are limited to a few isolated nunataks making it difficult to assess the long-term deglacial history of the area. To address this, we correlate existing records of lateral ice-stream retreat from marine sediment cores with onshore glacial thinning in two key areas of eastern Marie Byrd Land: the Kohler Range and Pine Island Bay. Our 10Be surface-exposure ages are the first from the isolated Kohler Range and show that the nunataks there became ice-free between 8.6 and 12.6 ka. This implies a minimum long-term average thinning rate of 3.3 ± 0.3 cm/yr, which is one order of magnitude lower than recent rates based on satellite data. We also present pre- to early Holocene 10Be surface-exposure ages from two islands located approximately 80 km downstream of the Pine Island Glacier ice-shelf front to constrain the lateral deglacial history in the Pine Island Bay area. This study provides insight into the significance of local ice sheet variations and suggests that the post-LGM history in the Amundsen Sea sector was characterized by glacial thinning as well as lateral retreat in pre- to early Holocene times.

The Indus River, one of Asia’s premier rivers, drains the western Tibetan Plateau and the Nanga Parbat syntaxis. These two areas juxtapose some of the lowest and highest topographic relief and commensurate denudation rates in the Himalaya-Tibet orogen, respectively, yet the spatial pattern of denudation rates upstream of the syntaxis remains largely unclear, as does the way in which major rivers drive headward incision into the Tibetan Plateau. We report a new inventory of ^(10)Be-based basinwide denudation rates from 33 tributaries flanking the Indus River along a 320 km reach across the western Tibetan Plateau margin. We find that denudation rates of up to 110 mm k.y.^(–1) in the Ladakh and Zanskar Ranges systematically decrease eastward to 10 mm k.y.^(–1) toward the Tibetan Plateau. Independent results from bulk petrographic and heavy mineral analyses support this denudation gradient. Assuming that incision along the Indus exerts the base-level control on tributary denudation rates, our data show a systematic eastward decrease of landscape downwearing, reaching its minimum on the Tibetan Plateau. In contrast, denudation rates increase rapidly 150–200 km downstream of a distinct knickpoint that marks the Tibetan Plateau margin in the Indus River longitudinal profile. We infer that any vigorous headward incision and any accompanying erosional waves into the interior of the plateau mostly concerned reaches well below this plateau margin. Moreover, reported long-term (>10^6 yr) exhumation rates from low-temperature chronometry of 0.1–0.75 mm yr^(–1) consistently exceed our ^(10)Be-derived denudation rates. With averaging time scales of 10^3–10^4 yr for our denudation data, we report postglacial rates of downwearing in a tectonically idle landscape. To counterbalance this apparent mismatch, denudation rates must have been higher in the Quaternary during glacial-interglacial intervals.

Some of the largest and most erosive floods on Earth result from the failure of glacial dams. While potentially cataclysmic ice dams are recognized to have repeatedly formed along ice-sheet margins, much less is known about the frequency and longevity of ice dams caused by mountain glaciers, and their impact on landscape evolution. Here we present field observations and results from cosmogenic nuclide dating that allow reconstructing a > 100-k.y.-long history of glacial damming in the Shyok Valley, eastern Karakoram (South Asia). Our field observations provide evidence that Asia's second-longest glacier, the Siachen, once extended for over 180 km and blocked the Shyok River during the penultimate glacial period, leading to upstream deposition of a more than 400-m-thick fluvio-lacustrine valley fill. Be-10-depth profile modeling indicates that glacial damming ended with the onset of the Eemian interglacial and that the Shyok River subsequently incised the valley fill at an average rate of similar to 4-7 m k.y.(-1). Comparison with contemporary ice-dammed lakes in the Karakoram and elsewhere suggests recurring outburst floods during the aggradation period, while over 25 cycles of fining-upward lake deposits within the valley fill indicate impounding of floods from farther upstream. Despite prolonged damming, the net effect of this and probably earlier damming episodes by the Siachen Glacier is dominated by glacial erosion in excess of fluvial incision, as evidenced by a pronounced overdeepening that follows the glaciated valley reach. Strikingly similar overdeepened valleys at all major confluences of the Shyok and Indus Rivers with Karakoram tributaries indicate that glacial dams and subsequent outburst floods have been widespread and frequent in this region during the Quaternary. Our study suggests that the interaction of Karakoram glaciers with the Shyok and Indus Rivers promoted valley incision and headward erosion into the western margin of the Tibetan Plateau.

[1] High relief and steep topography are thought to result in high erosion rates. In the Rwenzori Mountains of the Albert Rift, East Africa, where more than 3 km of relief have formed during uplift of the Rwenzori fault block, overall low denudation rates prevail. We measured in situ-derived cosmogenic denudation rates of 28.2 to 131 mm/kyr in mountainous catchments, and rates of 7.8 to 17.7 mm/kyr on the adjacent low-relief East African Plateau. These rates are roughly an order of magnitude lower than in other settings of similar relief. We present an extensive geomorphological analysis, and find that denudation rates are positively correlated with relief, hillslope gradient, and channel steepness, indicating that river incision controls erosional processes. In most upper headwater reaches above Quaternary ELA levels (>4500 m a.s.l.), glacial imprinting, inherited from several older and recent minor glaciation stages, prevails. In regions below 4500 m a.s.l., however, mild climatic conditions impede frost shattering, favor dense vegetation, and minimize bare rock areas and associated mass wasting. We conclude that erosion of the Rwenzori Mountains is significantly slower than corresponding rates in other mountains of high relief, due to a combination of factors: extremely dense mountain cloud forest vegetation, high rock strength of gneiss and amphibolite lithologies, and low internal fracturing due to the extensional tectonic setting. This specific combination, unique to this extensional tropical setting, leads to unexpected low erosion rates that cannot outpace post-Pliocene ongoing rock uplift of the Rwenzori fault block.

Uplifted Neogene marine sediments and Quaternary fluvial terraces in the Mut Basin, southern Turkey, reveal a detailed history of surface uplift of the southern margin of the Central Anatolian plateau from the Late Miocene to the present. New surface exposure ages (10Be, 26Al and 21Ne) of gravels above fluvial strath terraces located between 28 and 135 m above the modern Göksu River in the Mut Basin yield ages ranging from ca. 25 to 130 ka, or an average incision rate of 0.52 to 0.67 mm/yr. Published biostratigraphic data together with new interpretations of the fossil assemblages from uplifted marine sediments reveal average uplift rates of 0.25 to 0.37 mm/yr since Late Miocene time (starting between 8 and 5.45 Ma), and 0.72 to 0.74 mm/yr after 1.62 to 1.66 Ma. Together with the terrace ages, the data imply 0.6 to 0.7 mm/yr surface uplift rates from 1.6 Ma to the present. The difference between the post-Late Miocene and post-1.6 Ma surface uplift rates can be reconciled by increasing uplift rates through time (though they have been relatively steady since 1.6 Ma), or by multi-phased uplift with slow uplift or subsidence in between. Longitudinal profiles of rivers in the upper catchment of the Mut-Ermenek Basin show no apparent lithologic or fault control on knickpoints that occur at 1.2 to 1.5 km elevation, implying a transient response to a change in uplift rates. Projections of graded upper relict channel segments to the modern outlet, together with constraints from uplifted marine sediments show that a slower incision/uplift rate of ca. 0.1 mm/yr preceded the 0.7 mm/yr uplift rate. The river morphology and profile projections therefore suggest that the southern margin of the plateau was characterized by multi-phased uplift, rather than by steadily increasing uplift rates. Multi-phased uplift can be explained by surface uplift above a slab break-off and possibly also the arrival of the Eratosthenes Seamount at the collision zone south of Cyprus.

Rivers play an important role in continental erosion as they are the primary agents of transferring erosion products to the ocean. Understanding rivers and their transport pathways will improve the perception of many processes of global significance, such as biogeochemical cycling of pollutants and nutrients, atmospheric CO2 drawdown, soil formation and their erosion, crust evolutionin short the interaction between the atmospheric and the lithospheric compartment of the Earth s system (Allen, 2008). This interaction is characterised by the relative proportions of mechanical degradation vs. chemical weathering, whose products are, in dissolved or solid form, transported by rivers. The sediment load of rivers is thereby controlled by catchment relief, the channel slope and its connectivity to the hill slope, but also by climatic factors such as precipitation. The latter, together with temperature, exert control over chemical weathering that is dependent on physical erosion to a degree that is yet unknown (Anderson et al., 2002; Gaillardet et al., 1999; Riebe et al., 2001). Both mechanical erosion and chemical weathering, are, however, governed by tectonic activity, which drives processes of landscape rejuvenation and preconditions the fluvial transport regime (von Blanckenburg et al., 2004). On the shorter time scale, humans may act as geomorphic agents by constructing dams and reservoirs, and changing land use by deforestation and mining (Hooke, 2000; Syvitski et al., 2005; Wilkinson and McElroy, 2007). In tropical regions around the globe, large river basins are especially concentrated, and their behaviour plays an important role in river sediment transport (Latrubesse et al., 2005). For example, the tropics represent 25% of the total continental lands and contain 57% of the

We use cosmogenic nuclide-derived denudation rates from in situ-produced Be-10 in river sediment to determine sediment production rates for the central Amazon River and its major tributaries. Recent developments have shown that this method allows calculating denudation rates in large depositional basins despite intermediate sediment storage, with the result that fluxes of the sediment-producing hinterland can now be linked to those discharged at the basins' outlet. In rivers of the central Amazonian plain, sediment of finer grain sizes (125-500 mu m) yields a weighted cosmogenic nuclide-derived denudation rate of 0.24 +/- 0.02 mm/yr that is comparable to the integrated rate of all main Andean-draining rivers (0.37 +/- 0.06 mm/yr), which are the Beni, Napo, Mamore, Ucayali, and Maranon rivers. Coarser-grained sediment (> 500 mu m) of central Amazonian rivers is indicative of a source from the tectonically stable cratonic headwaters of the Guyana and Brazilian shields, for which the denudation rate is 0.01-0.02 mm/yr. Respective sediment loads can be calculated by converting these cosmogenic nuclide-derived rates using their sediment-producing areas. For the Amazon River at Obidos, a sediment production rate of similar to 610 Mt/yr results; non-Andean source areas contribute only similar to 45 Mt/yr. A comparison with published modern sediment fluxes shows similarities within a factor of similar to 2 with an average gauging-derived sediment load of similar to 1000 Mt/yr at Obidos, for example. We attribute this similar trend in cosmogenic versus modern sediment loads first to the absence of long-term deposition within the basin and second to the buffering capability of the large Amazon floodplain. The buffering capability dampens short-term, high-amplitude fluctuations (climatic variability in source areas and anthropogenic soil erosion) by the time the denudation rate signal of the hinterland is transmitted to the outlet of the basin.

Cosmogenic nuclide-based denudation rates and published erosion rates from recent river gauging in the Napo River basin (Peruvian Amazonia) are used to decipher erosion and sedimentation processes along a 600 km long transect from the headwaters to the lowlands. The sediment-producing headwaters to the Napo floodplain are the volcanically active Ecuadorian Andes, which discharge sediment at a cosmogenic nuclide-based denudation rate of 0.49 ± 0.12 mm/yr. This denudation rate was calculated from an average 10Be nuclide concentration of 2.2 ± 0.5 × 104 at/g(Qz) that was measured in bedload-derived quartz. Within the Napo lowlands, a significant drop in trunk stream 10Be nuclide concentrations relative to the Andean hinterland is recorded, with an average concentration of 1.2 ± 0.5 × 104 at/g(Qz). This nuclide concentration represents a mixture between the 10Be nuclide concentration of eroded floodplain deposits, and that of sediment eroded from the Andean hinterland that is now carried in the trunk stream. Evidence for addition of sediment from the floodplain to the trunk stream is provided by published decadal-scale sediment flux measurements from gauging stations operated in the Napo basin, from which an increase from 12 × 106 t/yr at the outflow of the Andes to not, vert, similar 47 × 106 t/yr at the confluence with the Solimões (upper Amazon River) is recorded. Therefore, approximately 35 × 106 t of floodplain sediment are added annually to the active Napo trunk stream. Combined with our nuclide concentration measurements, we can estimate that the eroded floodplain deposits yield a nuclide concentration of not, vert, similar 0.9 × 104 at/g(Qz) only. Under steady state surface erosion conditions, this nuclide concentration would translate to a denudation rate of the floodplain of not, vert, similar 0.47 mm/yr. However, we have no geomorphologic explanation for this high denudation rate within the low relief floodplain and thus suggest that this low-nuclide concentrated sediment is Andean-derived and would have been deposited in the floodplain at a time when erosion rates of the Andes were elevated. Therefore, the recently eroded floodplain sediment provides an Andean “paleo denudation rate” of 1.2 mm/yr that was calculated for high Andean production rates. A likely period for elevated erosion rates is the LGM, where climate and vegetation cover of the Andes differed from that of the Holocene. A possible cause for the erosion of the floodplain is the tectonic uplift of the Eastern Andes, which progressively shifts the Napo River northwards. Hence, the river cuts into ancient lowland sediment, which is admixed to the Andean sediment carried in the main Napo River.

Cosmogenic nuclides produced in quartz may either decay or accumulate while sediment is moved through a river basin. A change in nuclide concentration resulting from storage in a floodplain is potentially important in large drainage basins in which sediment is prone to repeated burial and remobilization as a river migrates through its floodplain. We have modeled depth- and time-dependent cosmogenic nuclide concentration changes for 10 Be, 26 Al, and 14 C during sediment storage and mixing in various active floodplain settings ranging from confined, shallow rivers with small floodplains to foreland-basin scale floodplains traversed by deep rivers. Floodplain storage time, estimated from channel migration rates, ranges from 0.4 kyr for the Beni River basin (Bolivia) to 7 kyr for the Amazon River basin, while floodplain storage depth, estimated from channel depth, ranges from 1 to 25 m. For all modeled active floodplain settings, the long-lived nuclides 10 Be and 26 Al show neither significant increase in nuclide concentration from irradiation nor decrease from decay. We predict a hypothetical response time after which changes in 10 Be or 26 Al concentrations become analytically resolvable. This interval ranges from 0.07 to 2 Myr and exceeds in all cases the typical residence time of sediment in a floodplain. Due to the much shorter half life of 14 C, nuclide concentrations modeled for the in situ-produced variety of this nuclide are, however, sensitive to floodplain storage on residence times of The cosmogenic nuclide composition of old deposits in currently inactive floodplains that have been isolated for periods of millions of years from the river that once deposited them is predicted to either increase or decrease in 10 Be and 26 Al concentration, depending on the depositional depth. These conditions can be evaluated using the 26 Al/ 10 Be ratio that readily discloses the depth and duration of storage. We illustrate these models with examples from the Amazon basin. As predicted, modern bedload collected from an Amazon tributary, the Bolivian Beni River, shows no systematic change in nuclide concentration as sediment is moved through 500 km of floodplain by river meandering. In contrast, in the central Amazon floodplain currently untouched by the modern river system, low 26 Al/ 10 Be ratios account for minimum burial depths of 5 to 10 m for a duration of > 5 Myr. The important result of this analysis is that in all likely cases of active floodplains, cosmogenic 10 Be and 26 Al concentrations remain virtually unchanged over the interval sediment usually spends in the basin. Thus, spatially-averaged denudation rates of the sediment-producing area can be inferred throughout the entire basin, provided that nuclide production rates are scaled for the altitudes of the sediment-producing area only, because floodplain storage does not modify nuclide concentrations introduced from the sediment source area.

[1] A north-south traverse through the Swiss Central Alps reveals that denudation rates correlate with recent rock uplift rates in both magnitude and spatial distribution. This result emerges from a study of in situ–produced cosmogenic 10Be in riverborne quartz in Central Alpine catchments. As a prerequisite, we took care to investigate the potential influence of shielding from cosmic rays due to snow, glaciers, and topographic obstructions; to calculate a possible memory from Last Glacial Maximum (LGM) glaciation; and to identify a watershed size that is appropriate for systematic sampling. Mean denudation rates are 0.27 ± 0.14 mm/a for the Alpine foreland and 0.9 ± 0.3 mm/a for the crystalline Central Alps. The measured cosmogenic nuclide-derived denudation rates are in good agreement with post-LGM lake infill rates and are about twice as high as denudation rates from apatite fission track ages that record denudation from 9 to 5 Ma. In general, denudation rates are high in areas of high topography and high crustal thickness. The similarity in the spatial distribution and magnitude of denudation rates and those of rock uplift rates can be interpreted in several ways: (1) Postglacial rebound or climate change has introduced a transient change in which both uplift and denudation follow each other with a short lag time; (2) the amplitude of glacial to interglacial changes in both is small and is contained in the scatter of the data; (3) both are driven by ongoing convergence where their similarity might hint at some form of long-term quasi steady state; or (4) enhanced continuous Quaternary erosion and isostatic compensation of the mass removed accounts for the distribution of present-day rock uplift.