Cosmogenic surface exposure dating the last deglaciation in Denmark: Discrepancies with independent age constraints suggest delayed periglacial landform stabilisation

Cosmogenic nuclide surface exposure ages are determined from in situ ¹⁰Be and ³⁶Cl analysis of 38 rock surfaces found in different glacial landforms in Denmark. Dating of erratic boulders and adjacent ice-sculpted bedrock on the island of Bornholm in the western Baltic Sea reveals almost identical values. This suggests that little if any inherited nuclides are present in the sampled boulders. West of the Last Glacial Maximum (LGM) ice margin in Denmark ages reflect exposure from the Middle Weichselian. East of the LGM margin exposure ages from 35 samples show Late Weichselian ages in a range between 20.6–11.9 ka. To test to what extent these dates reflect the onset of deglaciation immediately after cessation of active glacier flow, surface exposure ages are evaluated against independent chronologies of Late Weichselian ice-sheet fluctuations in southwestern Scandinavia. The Bornholm dates agree with the independent age model, however, in the data set for eastern Denmark only less than half the surface exposure ages lie within the expected age envelope. This apparent mismatch is most likely due to post-glaciation shielding and delayed surface stabilisation compared to the timing of ice-margin retreat. Thus ages from boulders resting in dead-ice moraines and mass wasting landscapes underestimate deglaciation by 3–6 thousand years. The results quantify the impact of exhumation and landform stabilisation on cosmogenic surface exposure ages on millennial scales. We conclude, that interpretation of cosmogenic exposure ages should include careful evaluation of possible post-depositional landform transformation in attempts to fine tune ages of e.g. end moraine features. With reference to independent age models we critically evaluate glacier advance – retreat scenarios from regions around the southern Baltic that alone are based on weighted average ages of cosmogenic exposure dating.     

Quaternary ice thinning of David Glacier in the Terra Nova Bay region,Antarctica

Understanding the history of Antarctic glaciation is important for interpreting paleoclimatic changes and estimating the changes in climate, sea level, and ice volume in the future. Ice core studies of the East Antarctic Ice Sheet (EAIS) and marine sediment cores from the entire Ross Sea have employed numerous proxies to reconstruct the glacial history of the Antarctic region. However, the ice and marine core records can be biased because of their specific locations, such as the uppermost accumulation zone or the terminus of the ablation zone, thereby introducing significant uncertainties in ice modeling. In this study, we analyzed 34 new ¹⁰Be and ²⁶Al samples from four benches that were glaciated in the past by David glacier and incorporate the present ice-free flat surfaces. We suggest that the David glacier experienced monotonic and stepwise vertical lowering along the flanks of Mt. Priestley since the early Pleistocene. The uppermost bedrock benches on Mt. Priestley were exposed at 1.77 ± 0.32 Ma, with no evidence of subsequent overriding by readvancing ice. At Mt. Priestley, the David glacier has been characterized by a cold-based regime since 1.77 Ma, with a denudation rate of only ∼16 cm/Ma, corresponding to the regional transition from warm to cold-based glaciation at 3.5 Ma. Simple exposure ages from two lower benches date to Marine Isotope Stage (MIS) 7 (234.1 ± 13.1 ka; 545 m asl) and MIS 4 (64.8 ± 13.7 ka; 222 m asl), suggesting that, since MIS 8, the overall lowering of glaciers has remained monotonic. The upper bench marks the lower limit of the MIS 8 glacial period and the upper limit of Penultimate Glacial Maximum (MIS 6), while the lower landform defines the upper limit of the last glacial period (MIS 4–2). The magnitude of Quaternary ice thinning at the David glacier was the highest (∼990 m) in the present terminal area (i.e., the most sensitive ablation zone), in contrast to the other outlet glaciers draining into the Terra Nova Bay, which experienced less ice lowering. The combination of the terrestrial (in situ ¹⁰Be and ²⁶Al) and previous marine (authigenic ¹⁰Be) cosmogenic data used in our study document the history of lowering of the David glacier driven by climatic changes during the Pleistocene. Both deglaciation and glaciation were limited during the mid-Pleistocene transition (MPT) and prior to the mid-Bruhnes event (MBE), due to the prevailing cold and arid climate, whereas deglaciation was dominant during other warm periods.     

Deglaciation history at the Alpine-Mediterranean transition (Argentera-Mercantour,SW Alps) from 10Be dating of moraines and glacially polished bedrock

Estimating the extent and age of the last glacial maxima as well as the chronology of glacial recessions in various environmental contexts is key to source-to-sink studies and paleoclimate reconstructions. The Argentera-Mercantour massif is located at the transition between the Alps and the Mediterranean Sea, therefore, its deglaciation chronology can be compared to the sediment budget of the Var River basin. Based on 13 new cosmic-ray exposure (CRE) beryllium-10 (¹⁰Be) datings performed on moraines and polished crystalline bedrocks and 22 reassessed ¹⁰Be CRE ages from similar altitude nearby steep basement surfaces, and from a lake sediment core, we can constrain the deglaciation chronology of the Argentera-Mercantour massif. These data allow for the first time to fully reconstruct the deglaciation history at the scale of the entire massif in agreement with a major glacier recession at c. 15 ka, at the onset of Bølling transition between the Oldest and Older Dryas. Main deglaciation of the upper slopes [2700–2800 m above sea level (a.s.l.)] occurred after the Last Glacial Maximum (LGM) at 20.8–18.6 ka, followed by the main deglaciation of the lower slopes (2300 m a.s.l.) at 15.3–14.2 ka. Finally, the flat polished surfaces above 2600 m a.s.l. and the zones confined within narrow lateral valleys were likely affected by progressive ice melting of remaining debris covered glaciers and moraine erosion following the Younger Dryas re-advance stage between 12 and 8–9 ka. At lower elevations, the Vens Lake located at 2300 m a.s.l., allows evidence of the onset of lake sedimentation at c. 14 ka and a transition towards a vegetated environment that mainly occurred before 8 ka. Moraine final stabilization at 5 ka might reflect denudation acceleration during the Holocene humid phase. This contribution reveals a glacier–climate relationship more sensitive to warming phases in the southern Alps highlighted by a major decrease of glaciers after c. 15 ka. This major deglaciation is correlated with a 2.5-fold decrease of sediment discharge of rivers into the Mediterranean Sea. © 2019 John Wiley & Sons, Ltd.     

10Be surface-exposure age dating of the Last Glacial Maximum in the northern Pamir (Tajikistan)

Knowledge of the spatial and temporal variations in Alpine glaciations is essential for reconstructing the regional and global timing of ice ages. This study investigates glacial deposits at the mouth of the Muksu catchment in the northern Pamir using ¹⁰Be surface-exposure age dating. We sampled boulders from the furthest downstream recessional moraine (20 samples) and five lateral moraines (41 samples) near the former terminus of the Fedchenko Glacier, the longest (∼72 km) present-day Alpine glacier of the Pamir. After the identification of outliers, the boulder population of the recessional moraine yielded a mean exposure age of 17.5 ± 1.9 ka. The maximum exposure age of the lateral moraines, collected ∼5 km up-valley of the recessional moraine, is 18.2 ± 1.7 ka. The boulder ages reflect glacial deposition during the Last Glacial Maximum (Marine Isotope Stage 2) in the region; they are in accordance with published glacial deposition ages in the western Tian Shan.     

Cosmogenic 36Cl glacial chronologies of the Late Quaternary glaciers on Mount Geyikdağ in the Eastern Mediterranean

We report the timing of glaciations during the Late Quaternary in the central Taurus Mountains of Turkey in the Eastern Mediterranean. Forty moraine samples from three glacial valleys on Mount Geyikdağ (36.53°N, 32.10^°E, 2877 m), near the Eastern Mediterranean coast of Turkey, were dated with in-situ cosmogenic ³⁶Cl. These glacial valleys are located on the southern flank of the mountain and were filled with few km long glaciers that terminated at elevations of about 1750 m above sea level. Three glacial retreats/advances were determined in this study. During the Last Glacial Maximum (LGM), glaciers reached their maximum positions at 20.6 ± 0.6 ka ago (±1σ). This date is in accordance with the timing of local maximum ice extent, represented by piedmont glaciers in the northern side of the mountain. Glaciers started to retreat after the LGM and shortly stabilized or re-advanced two times before they completely vanished out. The first stage ended before 13.7 ± 0.8 ka ago during the Late-glacial. The last glaciation occurred during the Holocene and ended between 9.6 ± 1.4 ka and 5.9 ± 0.5 ka ago. Later, glaciers mostly vanished from the study area, but a few rock glaciers developed during the Late Holocene. Glacial chronologies of Mount Geyikdağ are mostly comparable with the globally observed advances elsewhere.     

Can in-situ cosmogenic 14C be used to assess the influence of clast recycling on exposure dating of ice retreat in Antarctica?

Cosmogenic nuclide exposure dating of glacial clasts is becoming a common and robust method for reconstructing the history of glaciers and ice sheets. In Antarctica, however, many samples exhibit cosmogenic nuclide ‘inheritance’ as a result of sediment recycling and exposure to cosmic radiation during previous ice free periods. In-situ cosmogenic ¹⁴C, in combination with longer lived nuclides such as ¹⁰Be, can be used to detect inheritance because the relatively short half-life of ¹⁴C means that in-situ ¹⁴C acquired in exposure during previous interglacials decays away while the sample locality is covered by ice during the subsequent glacial. Measurements of in-situ ¹⁴C in clasts from the last deglaciation of the Framnes Mountains in East Antarctica provide deglaciation ages that are concordant with existing ²⁶Al and ¹⁰Be ages, suggesting that in this area, the younger population of erratics contain limited inheritance.     

Micro-hole and multigrain quartz luminescence dating of Paleodeltas at Lake Fryxell,McMurdo Dry Valleys (Antarctica), and relevance for lake history

Relict (perched) lacustrine deltas around the perennially ice-covered lakes in the Taylor Valley, Antarctica, imply that these lakes were up to 40 times larger in area than at present since the last glacial maximum (LGM). These deltas have been used to constrain ice-margin positions in Taylor Valley, and the boundaries of the proposed LGM ice-damned Glacial Lake Washburn. The timing of these high lake levels has depended on ¹⁴C chronologies of algal layers within relict lacustrine deltas. To provide additional geochronometric data for the post-LGM lake-level history, we applied photon-stimulated-luminescence (PSL) sediment dating to polymineral fine silt and sand-size quartz from 7 perched-delta and 3 active-delta sites of different elevations along 3 major meltwater streams entering Lake Fryxell. Our PSL dating of 4 quartz-sand samples from core tops in the seasonal ice-free moat of Lake Fryxell (elevation ∼18 m a.s.l.) and two core-top moat samples from the seasonal moat of Lake Vanda in nearby Wright Valley establish that adequate PSL clock zeroing (by daylight) occurs in regional, modern shoreline deposits. Minimum-age micro-hole PSL results from the moats are consistently near 100 a. Minimum-age micro-hole age estimates for the deltas range from ∼50 to 100 a near the present lake level up to 13.4 ± 1.3 ka at 240 m. These are systematically younger than the comparable, reservoir-uncorrected, ¹⁴C ages that range from 7 ka (cal yr BP) to 13 ka (cal yr BP) near lake level up to 20 ka (cal yr BP) at 220–240 m elevation. Our results indicate the occurrence of a dramatic discrepancy between PSL minimum-age and ¹⁴C age estimates that is presently unresolved.     

Glacier response to the change in atmospheric circulation in the eastern Mediterranean during the Last Glacial Maximum

In this study, we document glacial deposits and reconstruct the glacial history in the Karagöl valley system in the eastern Uludağ in northwestern Turkey based on 42 cosmogenic ¹⁰Be exposure ages from boulders and bedrock. Our results suggest the Last Glacial Maximum (LGM) advance prior to 20.4 ± 1.2 ka and at least three re-advances until 18.6 ± 1.2 ka during the global LGM within Marine Isotope Stage-2. In addition, two older advances of unknown age are geomorphologically well constrained, but not dated due to the absence of suitable boulders. Glaciers advanced again two times during the Lateglacial. The older is exposure dated to not later than 15.9 ± 1.1 ka and the younger is attributed to the Younger Dryas (YD) based on field evidence. The timing of the glaciations in the Karagöl valley correlates well with documented archives in the Anatolian and Mediterranean mountains and the Alps. These glacier fluctuations may be explained by the change in the atmospheric circulation pattern during the different phases of North Atlantic Oscillation (NAO) winter indices.     

Luminescence chronology of aeolian deposits from the Qinghai Lake area in the Northeastern Qinghai-Tibetan Plateau and its palaeoenvironmental implications

Qinghai Lake, on the northeastern Qinghai-Tibetan Plateau, is the largest extant closed-basin lake in China, and has been the subject of numerous palaeoclimatological and palaeoenvironmental studies. In this study, 32 samples of aeolian sand, loess and palaeosol at six sites, and 1 sample of shoreline deposits underlying aeolian deposits were dated using optically stimulated luminescence (OSL). Where available, OSL ages are in agreement with previously published ¹⁴C ages. Our dating results, in combination with previous published ages on aeolian deposits showed that: (1) The oldest aeolian deposits around Qinghai Lake are in excess of 165 ka. (2) Aeolian deposition then began at ∼14 ka in the Qinghai Lake area. Periods of palaeosol formation occurred at ∼16.9 ka, ∼12.2–11 ka, ∼10–9 ka, ∼5.2–4 ka, and ∼3.9–0.7 ka. (3) The accumulation intervals of palaeosols are generally consistent with drilling-core-based environmental change proxies, indicating that palaeosols were formed during wet periods with higher vegetation cover. (4) A depositional hiatus period of ∼40–50 ka exists between the surface mantle aeolian deposits and underlying gravel deposits. (5) Lake levels during the Holocene did not exceed 3205.2 m elevation (11.8 m above recent lake level of April, 2010).     

Applying a Bayesian approach for refining the chronostratigraphy of the Yumidong site in the Three Gorges region,central China

Yumidong (Corn Cave) is a newly discovered Paleolithic site in the Three Gorges region of central China. Numerous Paleolithic artifacts have been excavated from the sedimentary deposits of the cave in association with faunal remains attributed to the Middle-Late Pleistocene Ailuropoda-Stegodon fauna of southern China. To establish the chronology of the sedimentary sequence (>5 m thick), ¹⁴C dating was applied to bone and charcoal samples (n = 6); the U-series method was used to date in situ precipitated speleothems (n = 12), transported speleothem samples (n = 6) and 18 subsamples of a fossil tooth; and the coupled ESR/U-series method was used to date fossil teeth (n = 6). The derived dates were combined using a hierarchical Bayesian approach to generate a unified chronostratigraphy for the Yumidong sequence. In our Bayesian analyses, the ¹⁴C and coupled ESR/U-series dates were considered to provide direct age estimates for the target layers, while the U-series dates of the in situ precipitated speleothems and fossil tooth were used as minimum age constraints and those of the transported speleothem fragments as maximum age constraints. The Bayesian analyses provided robust time intervals for the archeological layers: L2-Upper (14–23 ka), L2-Lower (27–63 ka), L3 (106–171 ka), L4 (140–192 ka), L10 (157–229 ka), L11 (181–256 ka), and L12 (214–274 ka) with a probability of 95%, allowing the establishment of a ∼300 ka long geological and archeological history for the Yumidong site and placing it as a reference site for Paleolithic cultural evolution in the Three Gorges region from the late Middle Pleistocene to Late Pleistocene.     

Multi-method pIRIR dating of sedimentary sequences at the southern edge of the Gurbantunggut Desert,NW China and its palaeoenvironmental implications

The Gurbantunggut Desert is the second-largest desert in China, located in the westerly-dominated region of north-western China. Previous understanding of palaeoclimate of this desert was mostly based on lake and loess records from the Junggar Basin and Tian Shan Mountains, whilst direct dating of sedimentary records from the desert was very limited. This study applies high-resolution post-infrared infrared stimulated luminescence (pIRIR) dating to three sedimentary profiles at the southern edge of the Gurbantunggut Desert, which contain aeolian sand and water-lain sediments, recording palaeoenvironmental changes at the desert margin since the Last Glacial Maximum (LGM). Different pIRIR dating procedures were applied for samples with different ages. For Holocene-aged samples, a single-aliquot regenerative-dose (SAR) pIRIR procedure based on a three-stepped pIRIR measurement at 110 °C, 140 °C and 170 °C was used, and a standard growth curve (SGC) procedure yields an equivalent dose (D_e) similar to that of the full-SAR procedure; thus, is applicable for accelerating D_e measurement. For samples much older than the Holocene, a multi-aliquot regenerative-dose (MAR) pIRIR procedure based on a three-stepped pIRIR measurement at 150 °C, 200 °C and 250 °C was found to be the optimal dating procedure, because a SAR procedure would yield underestimated ages due to uncorrected initial sensitivity change. pIRIR dating results of the investigated profiles reveal a substantial sand accumulation during the LGM, an intensification of aeolian deposition at ∼12 ka and a wetter depositional environment at ∼10–8 ka. A rapid fluvial deposition is dated at ∼20–19 ka, corresponding to the deglaciation period. The sedimentary records from the desert margin show some correlation with lake and loess records in the same region and suggest a complex response of the desert environment to different climatic factors.     

10Be dating of boulders on moraines from the last glacial period in the Nyainqentanglha mountains,Tibet

Chronologies of glacial advances during the last glacial period in the Nyainqentanglha mountain range may provide constraints on the past climate in a transition zone of the Asian monsoon.We present 15 new10Be exposure ages from two moraines in the Payuwang valley,on the north slope of the range.The inner moraine has exposure ages ranging from 18.0±1.7 to 30.6±2.8 ka(n=10),with a mean age of 23.8±4.0 ka,corresponding to the global Last Glacial Maximum(LGM).The outer moraine yields exposure ages ranging from 18.0±1.6 to 39.9±3.7 ka(n=5).Evidence for weathering leads us to view the oldest age as a minimum age,placing moraine formation during MIS3.Chronologies from the last glacial period from south slope of the Nyainqentanglha support this interpretation.Thus,there appears to have been a local LGM(LLGM)during MIS3 and a more limited glacial advance during the global LGM.Glacial advances during MIS3 in the Nyainqentanglha may correlate with millennial-scale climate change(Heinrich events).     

Ice cap erosion patterns from bedrock 10Be and 26Al,southeastern Tibetan Plateau

Quantifying glacial erosion contributes to our understanding of landscape evolution and topographic relief production in high altitude and high latitude areas. Combining in situ ¹⁰Be and ²⁶Al analysis of bedrock, boulder, and river sand samples, geomorphological mapping, and field investigations, we examine glacial erosion patterns of former ice caps in the Shaluli Shan of the southeastern Tibetan Plateau. The general landform pattern shows a zonal pattern of landscape modification produced by ice caps of up to 4000 km² during pre-LGM (Last Glacial Maximum) glaciations, while the dating results and landforms on the plateau surface imply that the LGM ice cap further modified the scoured terrain into different zones. Modeled glacial erosion depth of 0–0.38 m per 100 ka bedrock sample located close to the western margin of the LGM ice cap, indicates limited erosion prior to LGM and Late Glacial moraine deposition. A strong erosion zone exists proximal to the LGM ice cap marginal zone, indicated by modeled glacial erosion depth >2.23 m per 100 ka from bedrock samples. Modeled glacial erosion depths of 0–1.77 m per 100 ka from samples collected along the edge of a central upland, confirm the presence of a zone of intermediate erosion in-between the central upland and the strong erosion zone. Significant nuclide inheritance in river sand samples from basins on the scoured plateau surface also indicate restricted glacial erosion during the last glaciation. Our study, for the first time, shows clear evidence for preservation of glacial landforms formed during previous glaciations under non-erosive ice on the Tibetan Plateau. As patterns of glacial erosion intensity are largely driven by the basal thermal regime, our results confirm earlier inferences from geomorphology for a concentric basal thermal pattern for the Haizishan ice cap during the LGM. © 2018 John Wiley & Sons, Ltd.     

Surface exposure dating of Holocene basalt flows and cinder cones in the Kula volcanic field (Western Turkey) using cosmogenic 3He and 10Be

The Kula volcanic field in Western Turkey comprises about 80 cinder cones and associated basaltic lava flows of Quaternary age. Based on geomorphological criteria and K-Ar dating, three eruption phases, β2–β4, were distinguished in previous studies. Human footprints in ash deposits document that the early inhabitants of Anatolia were affected by the volcanic eruptions, but the age of the footprints has been poorly constrained. Here we use ³He and ¹⁰Be exposure dating of olivine phenocrysts and quartz-bearing xenoliths to determine the age of the youngest lava flows and cinder cones. In the western part of the volcanic field, two basalt samples from a 15-km-long block lava flow yielded ³He ages of 1.5 ± 0.3 ka and 2.5 ± 0.4 ka, respectively, with the latter being in good agreement with a ¹⁰Be age of 2.4 ± 0.3 ka for an augen gneiss xenolith from the same flow. A few kilometers farther north, a metasedimentary xenolith from the top of the cinder cone Çakallar Tepe gave a ¹⁰Be age of 11.2 ± 1.1 ka, which dates the last eruption of this cone and also the human footprints in the related ash deposits. In the center of the volcanic field, a basalt sample and a metasedimentary xenolith from another cinder cone gave consistent ³He and ¹⁰Be ages of 2.6 ± 0.4 ka and 2.6 ± 0.3 ka, respectively. Two β4 lava flows in the central and eastern part of the volcanic province yielded ³He ages of 3.3 ± 0.4 ka and 0.9 ± 0.2 ka, respectively. Finally, a relatively well-preserved β3 flow gave a ³He age of ∼13 ka. Taken together, our results demonstrate that the penultimate eruption phase β3 in the Kula volcanic field continued until ∼11 ka, whereas the youngest phase β4 started less than four thousand years ago and may continue in the future.     

Dating Holocene lavas on Stromboli,Italy using cosmogenic He

In-situ cosmogenic ³He exposure ages of pyroxene phenocrysts from basalts from the Upper Neostromboli formation in southwest Stromboli date its eruption at 7.0 ± 0.3 ka (1σ, n = 3, Ginostra site) and 6.8 ± 0.2 ka (1σ, n = 10, Timpone del Fuoco site) respectively. Correlation of our new data to previous K/Ar and palaeomagnetic ages from the northwestern Neostromboli phase suggests that it erupted within a confined period between roughly 6 and 14 ka. The low uncertainty on the ³He_cos ages as well as on individual exposure ages (4.4–8.7%) demonstrates that ³He_cos exposure dating is a viable tool for dating Holocene basalt lavas. The ages compare favourably to uncertainties obtained for radiocarbon dating of similar rocks.     

Fuerteventura – Assessment of a calibration site for cosmogenic 3He exposure dating with the 40Ar/39Ar incremental heating method

In situ Terrestrial Cosmogenic Nuclides (hereafter TCNs) are increasingly important for absolutely dating terrestrial events and processes. This study aimed at improving our knowledge of the production rate of Terrestrial Cosmogenic ³He formed in situ in rock surfaces at low latitude and sea level as well as re-evaluation of the Canary Islands as a calibration site for TCNs. For this purpose, we sampled basaltic lava flows from some of the youngest and yet undated volcanic sites and used the ⁴⁰Ar/³⁹Ar incremental heating method on groundmass samples and in situ cosmogenic ³He on olivine and clinopyroxene phenocrysts. ⁴⁰Ar/³⁹Ar analysis was done on a Hiden HAL Series 1000 triple filter quadrupole mass spectrometer with extraction furnace. Incremental heating data shows ages in the Late Pleistocene from 52.7 ± 21.6 ka to 398.6 ± 27.6 ka.We measured cosmogenic ³He concentrations in olivine and clinopyroxene phenocrysts from flow top samples on a MAP 215-50 sector mass spectrometer with a crushing device and a diode laser extraction system. Exposure age calculations yielded ages in the range 38.9 ± 4.0 ka to 62.3 ± 6.7 ka for the youngest lava flow and the data series is in broad agreement with the argon data up to 250 ka and reveals a more continuous time line of volcanism during the late Pleistocene on the island. However, the dataset was not sufficient for calculation of production rates for in situ Terrestrial Cosmogenic ³He as many samples showed signs of erosion. Calculated erosion rates range from none to as high as 7.3 mm/kyr assuming a rock density of 2.9 g/cm². This finding puts a constraint on the use of Fuerteventura as a calibration site for exposure histories older than 50–100 ka. A comparison with cosmogenic ³⁶Cl data supports these findings and indicates substantial weathering.     

Geochemical analyses of air from an ancient debris-covered glacier,Antarctica

We examined air trapped in ancient ice from three shallow cores (<35 m deep) recovered from stagnant portions of the Mullins glacier, an 8 km long debris-covered alpine glacier in the McMurdo Dry Valleys that is overlain by several in-situ volcanic ash-fall deposits. Previously reported ⁴⁰Ar/³⁹Ar dates on ash-fall in the vicinity of the core sites average 4.0 Ma, and underlying ice is presumably as old in some areas. We analyzed the elemental and isotopic composition of O₂, N₂, and Ar and total air content of the glacial ice. We also dated the trapped air directly to an uncertainty of ±220 kyr (1σ) by measuring its ⁴⁰Ar/³⁶Ar and ³⁸Ar/³⁶Ar ratios. Our results suggest that the air analyzed is likely a mixture of ancient atmosphere trapped at the time of ice formation and more recent air introduced via cracks in the ice that penetrate to at least 33 m. The isotopic signatures of gases have been complicated by gas loss, as well as a mixture of thermal and gravitational fractionation. The oldest age estimated for the trapped air dates to 1.6 Ma, indicating that the original air is at least as old as 1.6 ± 0.2 Ma. A convergence to older ice ages with increasing depth in the deepest core analyzed (33 m) hints at the possibility that pristine air might be recovered at greater depths. Minor interstitial debris present in the glacial ice (<1%), along with geochemical evidence for in-situ microbial respiration, prohibit direct analysis of CO₂. We measured the triple isotopic composition of O₂ as a proxy for CO₂ and infer that, in the air represented in our ice samples, CO₂ concentrations are within the range observed over the last 800 ka.     

A long-term rock uplift rate for eastern Crete and geodynamic implications for the Hellenic subduction zone

The island of Crete in the forearc of the Hellenic subduction zone has a rugged topography with local relief exceeding 2 km. Based on the elevation of marine shorelines, rates of rock uplift during the Late Holocene were previously estimated to range between 1 and 4 mm/a in different parts of the island. These rates may, however, not be representative for longer timescales, because subduction earthquakes with up to 9 m of vertical coseismic displacement have affected Crete in the Late Holocene. Here we use a well preserved sequence of marine terraces near Kato Zakros in eastern Crete to determine the rate of rock uplift over the last ∼600 ka. Field investigations and topographic profiles document a flight of more than 13 marine bedrock terraces that were carved into limestones of the Tripolitza unit. Preliminary age constraints for the terraces were obtained by ¹⁰Be exposure dating of rare quartz-bearing sandstone clasts, which are present on some terraces. The ¹⁰Be ages of these samples, which have been corrected for an inherited nuclide component, yielded exposure ages between ∼100 ka and zero. Combined with geomorphologic evidence the two oldest ¹⁰Be ages suggest that the terraces T₄ and T₅, with shoreline angles at an elevation of ∼68 and ∼76 m above sea level, respectively, formed during the marine isotope stage 5e about 120 ka ago. The correlation of the higher terraces (T₆ to T₁₃) with regional sea-level highstands indicates sustained rock uplift at a rate of ∼0.5 m/ka since at least ∼600 ka. As normal faulting has dominated the tectonics of Crete during the last several million years, upper crustal shortening can be ruled out as a cause for rock uplift. We argue that the sustained uplift of the island results from the continuous underplating of sediments, which are transferred from the subducting African plate to the base of the crust beneath Crete.     

Moraine pebbles and boulders yield indistinguishable Be ages: A case study from Colorado, USA

Cosmogenic exposure dating of moraines during the last two decades has vastly improved knowledge on the timing of glaciation worldwide. Due to a variety of geologic complications, such as moraine degradation, snow cover, bedrock erosion and isotopic inheritance, samples from multiple large boulders (>1–2 m) often lead to the most accurate moraine age assignments. However, in many cases, large boulders are not available on moraines of interest. Here, I test the suitability of pebble collections from moraine crest surfaces as a sample type for exposure dating. Twenty-two ¹⁰Be ages from two Pleistocene lateral moraine crests in Pine Creek valley in the upper Arkansas River basin, Colorado, were calculated from both pebble and boulder samples. Ten ¹⁰Be ages from a single-crested Bull Lake lateral moraine range between 3 and 72 ka, with no statistical difference between pebble (n = 5) and boulder (n = 5) ages. The lack of a cluster of ¹⁰Be ages suggests that moraine degradation has led to anomalously young exposure ages. Twelve ¹⁰Be ages from a single-crested Pinedale lateral moraine have a bimodal age distribution; one mode is 22.0 ± 1.4 ka (three boulders, two pebble collections), the other is 15.2 ± 0.9 ka (two boulders, five pebble collections). The interpretation of the two age modes is that two glacier maxima of similar extent were attained during the late Pleistocene. Regardless of moraine age interpretations, that ¹⁰Be ages from pebble collections and boulders are indistinguishable on moraines of two different ages, and in two different age modes of the Pinedale moraine, suggests that pebble collections from moraine crests may serve as a suitable sample type in some settings.     

The radiocarbon reservoir effect in proglacial lakes: Examples from Antarctica

We studied the apparently old radiocarbon ages from lakes in the dry valleys of Antarctica. The radiocarbon reservoir effect in these lakes results from two components: the inherited age and the residence age. The inherited age is derived from input of old carbon, primarily from subsurface melt of adjacent glaciers. The residence age comes from in situ aging of lake water in an environment sealed from the atmosphere. Our results indicate that surface melt of glaciers introduces little ancient carbon to the lake system, because of rapid gas equilibration with the atmosphere. Subsurface melt in lakes with large glacier cross-sectional areas at the grounding line, however, can contribute a significant amount of ancient carbon, leading to lake-bottom reservoir effects in excess of ∼ 2700 yr. This value can increase to ∼ 20,000 yr immediately at the grounding line. In most lakes, however, surface melt far exceeds that from the subsurface and dilutes the effect of ancient carbon, making the inherited age relatively low. Residence ages generally are on the order of a few thousand years, but can be as much as ∼ 10,000 yr. Because a residence age is reset when the lake loses its ice cover and is exposed to wind-driven mixing, its magnitude can provide important information about lake history.