Nature Geoscience

Drinking shallow groundwater with naturally elevated concentrations of arsenic is causing widespread disease in many parts of South and Southeast Asia. In the Bengal Basin, growing reliance on deep (>150 m) groundwater has lowered exposure. In the most affected districts of Bangladesh, shallow groundwater concentrations average 100 to 370 μg L(-1), while deep groundwater is typically < 10 μg L(-1). Groundwater flow simulations have suggested that, even when deep pumping is restricted to domestic use, deep groundwater in some areas of the Bengal Basin is at risk of contamination. However, these simulations have neglected the impedance of As migration by adsorption to aquifer sediments. Here we quantify for the first time As sorption on deeper sediments in situ by replicating the intrusion of shallow groundwater through injection of 1,000 L of deep groundwater modified with 200 μg L(-1) of As into a deeper aquifer. Arsenic concentrations in the injected water were reduced by 70% due to adsorption within a single day. Basin-scale modelling indicates that while As adsorption extends the sustainable use of deep groundwater, some areas remain vulnerable; these areas can be prioritized for management and monitoring.

Monomethylmercury is a neurotoxin that poses significant risks to human health1 due to its bioaccumulation in food webs. Sunlight degradation to inorganic mercury is an important component of the mercury cycle that maintains methylmercury at low concentrations in natural waters. Rates of photodecomposition, however, can vary drastically between surface waters2-5 for reasons that are largely unknown. Here, we show that photodegradation occurs through singlet oxygen, a highly reactive form of dissolved oxygen generated by sunlight irradiation of dissolved natural organic matter. The kinetics of degradation, however, depended on water constituents that bind methylmercury cations. Relatively fast degradation rates (similar to observations in freshwater lakes) applied only to methylmercury species bound to organic sulfur-containing thiol ligands such as glutathione, mercaptoacetate, and humics. In contrast, methylmercury-chloride complexes, which are dominant in marine systems, were unreactive. Binding by thiols lowered the excitation energy of the carbon-mercury bond on the methylmercury molecule6-7 and subsequently increased reactivity towards bond breakage and decomposition. Our results explain methylmercury photodecomposition rates that are relatively rapid in freshwater lakes2-4 and slow in marine waters5.

A minimum atmospheric temperature, or tropopause, occurs at a pressure of around 0.1 bar in the atmospheres of Earth, Titan, Jupiter, Saturn, Uranus and Neptune, despite great differences in atmospheric composition, gravity, internal heat and sunlight. In all these bodies, the tropopause separates a stratosphere with a temperature profile that is controlled by the absorption of shortwave solar radiation, from a region below characterised by convection, weather, and clouds. However, it is not obvious why the tropopause occurs at the specific pressure near 0.1 bar. Here we use a physically-based model to demonstrate that, at atmospheric pressures lower than 0.1 bar, transparency to thermal radiation allows shortwave heating to dominate, creating a stratosphere. At higher pressures, atmospheres become opaque to thermal radiation, causing temperatures to increase with depth and convection to ensue. A common dependence of infrared opacity on pressure, arising from the shared physics of molecular absorption, sets the 0.1 bar tropopause. We hypothesize that a tropopause at a pressure of approximately 0.1 bar is characteristic of many thick atmospheres, including exoplanets and exomoons in our galaxy and beyond. Judicious use of this rule could help constrain the atmospheric structure, and thus the surface environments and habitability, of exoplanets.

Decay of the CO2-dominated atmosphere is an important component of long-term environmental change on Mars, but direct constraints on paleoatmospheric pressure P are few. Of particular interest is the climate that allowed rivers to flow early in Mars history, which was affected by P via direct and indirect greenhouse effects. The size of craters embedded within ancient layered sediments is a proxy for P: the smaller the minimum-sized craters that form, the thinner the past atmosphere. Here we use high-resolution orthophotos and Digital Terrain Models (DTMs) to identify ancient craters among the river deposits of Aeolis close to Gale crater, and compare their sizes to models of atmospheric filtering of impactors by thicker atmospheres. We obtain an upper limit of P <= 760+/-70 mbar, rising to P <= 1640+/-180 mbar if rimmed circular mesas are excluded. Our work assumes target properties appropriate for desert alluvium: if sediment developed bedrock-like rock-mass strength by early diagenesis, the upper limit increases by a factor of up to 2. If Mars did not have a stable multibar atmosphere at the time that the rivers were flowing, the warm-wet CO2 greenhouse of Pollack et al. (1987) is ruled out, and long-term average temperatures were most likely below freezing.

As they keep cooling and contracting, Solar System giant planets radiate more energy than they receive from the Sun. Applying the first and second principles of thermodynamics, one can determine their cooling rate, luminosity, and temperature at a given age. Measurements of Saturn's infrared intrinsic luminosity, however, reveal that this planet is significantly brighter than predicted for its age. This excess luminosity is usually attributed to the immiscibility of helium in the hydrogen-rich envelope, leading to "rains" of helium-rich droplets. Existing evolution calculations, however, suggest that the energy released by this sedimentation process may not be sufficient to resolve the puzzle. Here, we demonstrate using planetary evolution models that the presence of layered convection in Saturn's interior, generated, like in some parts of Earth oceans, by the presence of a compositional gradient, significantly reduces its cooling. It can explain the planet's present luminosity for a wide range of configurations without invoking any additional source of energy. This suggests a revision of the conventional homogeneous adiabatic interior paradigm for giant planets, and questions our ability to assess their heavy element content. This reinforces the possibility for layered convection to help explaining the anomalously large observed radii of extrasolar giant planets.

Non-basaltic volcanism is rare on the Moon. The best known examples occur on the lunar nearside in the compositionally evolved Procellarum KREEP terrane. However, there is an isolated thorium-rich area—the Compton–Belkovich thorium anomaly—on the lunar farside for which the origin is enigmatic. Here we use images from the Lunar Reconnaissance Orbiter Cameras, digital terrain models and spectral data from the Diviner lunar radiometer to assess the morphology and composition of this region. We identify a central feature, 25 by 35 km across, that is characterized by elevated topography and relatively high reflectance. The topography includes a series of domes that range from less than 1 km to more than 6 km across, some with steeply sloping sides. We interpret these as volcanic domes formed from viscous lava. We also observe arcuate to irregular circular depressions, which we suggest result from collapse associated with volcanism. We find that the volcanic feature is also enriched in silica or alkali-feldspar, indicative of compositionally evolved, rhyolitic volcanic materials. We suggest that the Compton–Belkovich thorium anomaly represents a rare occurrence of non-basaltic volcanism on the lunar farside. We conclude that compositionally evolved volcanism did occur far removed from the Procellarum KREEP terrane.

Titan's equatorial regions are covered by eastward propagating linear dunes. This direction is opposite to mean surface winds simulated by Global Climate Models (GCMs), which are oriented westward at these latitudes, similar to trade winds on Earth. Different hypotheses have been proposed to address this apparent contradiction, involving Saturn's gravitational tides, large scale topography or wind statistics, but none of them can explain a global eastward dune propagation in the equatorial band. Here we analyse the impact of equinoctial tropical methane storms developing in the superrotating atmosphere (i.e. the eastward winds at high altitude) on Titan's dune orientation. Using mesoscale simulations of convective methane clouds with a GCM wind profile featuring superrotation, we show that Titan's storms should produce fast eastward gust fronts above the surface. Such gusts dominate the aeolian transport, allowing dunes to extend eastward. This analysis therefore suggests a coupling between superrotation, tropical methane storms and dune formation on Titan. Furthermore, together with GCM predictions and analogies to some terrestrial dune fields, this work provides a general framework explaining several major features of Titan's dunes: linear shape, eastward propagation and poleward divergence, and implies an equatorial origin of Titan's dune sand.

The Younger Dryas interval during the Last Glacial Termination was an abrupt return to glacial-like conditions punctuating the transition to a warmer, interglacial climate. Despite recent advances in the layer counting of ice-core records of the termination, the timing and length of the Younger Dryas remain controversial. Also, a steep rise in the concentration of atmospheric radiocarbon at the onset of the interval, recorded primarily in the Cariaco Basin, has been difficult to reconcile with simulations of the Younger Dryas carbon cycle. Here we discuss a radiocarbon chronology from a tree-ring record covering the Late Glacial period that has not been absolutely dated. We correlate the chronology to ice-core timescales using the common cosmic production signal in tree-ring 14C and ice-core 10Be concentrations. The results of this correlation suggest that the Cariaco record may be biased by changes in the concentration of radiocarbon in the upper ocean during the early phase of the Younger Dryas climate reversal in the Cariaco basin. This bias in the marine record may also affect the accuracy of a widely used radiocarbon calibration curve over this interval. Our tree-ring-based radiocarbon record is easily reconciled with simulated production rates and carbon-cycle changes associated with reduced ocean ventilation during the Younger Dryas.

During the last deglaciation, atmospheric carbon dioxide concentrations rose at the same time that the Delta14C of that CO2 fell. This has been attributed to the release of 14C-depleted carbon dioxide from the deep ocean, possibly vented through the Southern Ocean. Recently, a sediment record from the eastern North Pacific Ocean spanning the last deglaciation was interpreted to reflect transport of such radiocarbon-depleted CO2 from the Southern Ocean through Antarctic Intermediate Water. However, the suggestion that the record reflects intermediate water derived from the Southern Ocean remains controversial. Here we assess the source of the deglacial intermediate water by measuring the neodymium isotopes of fossil fish teeth/debris from the same eastern North Pacific core used in the earlier study. The isotopic signature of a water mass, which is captured in the fossil fish teeth, reflects the location in which it formed. Our data exhibit a clear shift in the neodymium isotope values towards Southern Ocean values about 18,000 years ago, coinciding with the negative Delta14C excursion. We conclude that these data support a Southern Ocean source for the deglacial radiocarbon-depleted CO2 detected in the eastern North Pacific.

Sea surface temperature variability in the North Atlantic Ocean recorded since about 1850 has been ascribed to a natural multidecadal oscillation superimposed on a background warming trend1-6. It has been suggested that the multidecadal variability may be a persistent feature6-8, raising the possibility that the associated climate impacts may be predictable7,8. owever, our understanding of the multidecadal ocean variability before the instrumental record is based on interpretations of high-latitude terrestrial proxy records. Here we present an absolutely dated and annually resolved record of sea surface temperature from the Bahamas, based on a 440-year time series of coral growth rates. The reconstruction indicates that temperatures were as warm as today from about 1552 to 1570, then cooled by about 1° C from 1650 to 1730 before warming until the present. Our estimates of background variability suggest that much of the warming since 1900 was driven by anthropogenic forcing. Interdecadal variability with a period of 15-25 years is superimposed on most of the record, but multidecadal variability becomes significant only after 1730. We conclude that the multidecadal variability in sea surface temperatures in the low-latitude western Atlantic Ocean may not be persistent, potentially making accurate decadal climate forecasts more difficult to achieve.

The North Atlantic Oscillation is a meridional oscillation of atmospheric mass measured between Iceland and the Açores, which drives winter climate variability in eastern North America and Europe. A prolonged period of the positive phase during the 1990s led to the suggestion that anthropogenic warming was affecting the behaviour of the North Atlantic Oscillation. However, instrumental records are too short to compare observations during periods of extended warm and cold hemispheric temperatures, and existing palaeoclimate reconstructions primarily capture terrestrial variability. Here we present a record of Sr/Ca, a sea surface temperature proxy, from a Bermuda coral from 1781 to 1999. We use this monthly resolved record to reconstruct past variability of the North Atlantic Oscillation at multiple frequencies. Our record shows enhanced multidecadal scale variability during the late twentieth century compared with the end of the Little Ice Age (1800-1850). We suggest that variability within the North Atlantic Oscillation is linked to the mean temperature of the Northern Hemisphere, which must be considered in any long-term predictions.

The largest earthquakes are generated in subduction zones, and the earthquake rupture typically extends for hundreds of kilometres along a single subducting plate. These ruptures often begin or end at structural boundaries on the overriding plate that are associated with the subduction of prominent bathymetric features of the downgoing plate. Here, we determine uplift and subsidence along shorelines for the 1 April 2007 moment magnitude MW 8.1 earthquake in the western Solomon Islands, using coral microatolls which provide precise measurements of vertical motions in locations where instrumental data are unavailable. We demonstrate that the 2007 earthquake ruptured across the subducting Simbo ridge transform and thus broke through a triple junction where the Australian and Woodlark plates subduct beneath the overriding Pacific plate. Previously, no known major megathrust rupture has involved two subducting plates. We conclude that this event illustrates the uncertainties of predicting the segmentation of subduction zone rupture on the basis of structural discontinuities.

Large, well-documented wildfires have recently generated worldwide attention, and raised concerns about the impacts of humans and climate change on wildfire regimes. However, comparatively little is known about the patterns and driving forces of global fire activity before the twentieth century. Here we compile sedimentary charcoal records spanning six continents to document trends in both natural and anthropogenic biomass burning for the past two millennia. We find that global biomass burning declined from AD 1 to 1750, before rising sharply between 1750 and 1870. Global burning then declined abruptly after 1870. The early decline in biomass burning occurred in concert with a global cooling trend and despite a rise in the human population. We suggest the subsequent rise was linked to increasing human influences, such as population growth and land-use changes. Our compilation suggests that the final decline occurred despite increasing air temperatures and population. We attribute this reduction in the amount of biomass burned over the past 150 years to the global expansion of intensive grazing, agriculture and fire management.

Southwestern Japan lies at the boundary between the subducting Philippine Sea plate and the overriding Eurasian plate. A magnitude 8 megathrust earthquake ruptured the Tonankai and Nankai segments in 1944 and 1946, respectively, but the neighbouring Tokai segment of the plate boundary remained locked1. A large megathrust earthquake in the Tokai region has therefore been expected. In 2009, a magnitude 6.4 earthquake took place in Suruga Bay, within the Philippine Sea subducting plate, close to the Tokai segment. Here, we use a fault-slip model to examine the impact of the stress changes2 caused by the Suruga Bay event on the Tokai segment. We show that the occurrence rate of plate-boundary seismicity increased following the earthquake. Most of the presumed strongly locked patches of the Tokai segment are located within areas of increased stress. Rupturing of a locked patch—following a threshold level of seismic stress—could trigger the rupture of the entire Tokai segment, leading to a megathrust earthquake.

Volcanoes at spreading centres on land often exhibit seismicity and ground inflation months to years before an eruption, caused by a gradual influx of magma to the source reservoir. Deflation and seismicity can occur on time scales of hours to days, and result from the injection of magma into adjacent rift zones. Volcanoes at submarine rift zones, such as Axial Seamount in the northeast Pacific Ocean, have exhibited similar behaviour, but a direct link between seismicity, seafloor deformation and magma intrusion has never been demonstrated. Here we present recordings from ocean-bottom hydrophones and an established array of bottom-pressure recorders that reveal patterns of both microearthquakes and seafloor deformation at Axial Seamount on the Juan de Fuca Ridge, before it erupted in April 2011. Our observations show that the rate of seismicity increased steadily during a period of several years, leading up to an intrusion and eruption of magma that began on 6 April 2011. We also detected a sudden increase in seismo-acoustic energy about 2.6h before the eruption began. Our data indicate that access to real-time seismic data, projected to be available in the near future, might facilitate short-term forecasting and provide sufficient lead-time to prepare in situ instrumentation before future intrusion and eruption events.

The Arctic climate is changing rapidly(1). From 1979 to 2006, September sea-ice extent decreased by almost 25% or about 100,000 km(2) per year (ref. 2). In September 2007, Arctic sea-ice extent reached its lowest level since satellite observations began(3) and in September 2008, sea-ice cover was still low. This development has raised concerns that the Arctic Ocean could be ice-free in late summer in only a few decades, with important economic and geopolitical implications. Unfortunately, most current climate models underestimate significantly the observed trend in Arctic sea-ice decline(4), leading to doubts regarding their projections for the timing of ice-free conditions. Here we analyse the simulated trends in past sea-ice cover in 18 state-of-art-climate models and find a direct relationship between the simulated evolution of September sea-ice cover over the twenty-first century and the magnitude of past trends in sea-ice cover. Using this relationship together with observed trends, we project the evolution of September sea-ice cover over the twenty-first century. We find that under a scenario with medium future greenhouse-gas emissions, the Arctic Ocean will probably be ice-free in September before the end of the twenty-first century.

Submarine groundwater discharge is defined as any flow of water at continental margins from the seabed to the coastal ocean, regardless of fluid composition or driving force1. The flux of submarine groundwater discharge has been hypothesized to be a pathway for enriching coastal waters in nutrients, carbon and metals2. Here, we estimate the submarine groundwater flux from the inventory of 228Ra in the upper Atlantic Ocean, obtained by interpolating measurements at over 150 stations. Only 46% of the loss in 228Ra from radioactive decay is replenished by input from dust, rivers and coastal sediments. We infer that the remainder must come from submarine groundwater discharge. Using estimates of 228Ra concentrations in submarine groundwater discharge, we arrive at a total flux from submarine groundwater discharge of 2–41013 m3 yr- 1, between 80 and 160% of the amount of freshwater entering the Atlantic Ocean from rivers. Submarine groundwater discharge is not a freshwater flux, but a flux of terrestrial and sea water that has penetrated permeable coastal sediments. Our assessment of the volume of submarine groundwater discharge confirms that this flux represents an important vehicle for the delivery of nutrients, carbon and metal to the ocean.

In the coming decades, significant changes in the polar regions will increase the contribution of ice sheets to global sea-level rise. Under the ice streams and outlet glaciers that deliver ice to the oceans, water and deformable wet sediments lubricate the base, facilitating fast ice flow. The influence of subglacial water on fast ice flow depends on the geometry and capacity of the subglacial hydrologic system: water moving rapidly through a well-connected system of conduits or channels will have little impact on ice-sheet velocities, but water injected into a spatially dispersed subglacial system may reduce the effective pressure at the base of the ice sheet, and thereby trigger increased ice-sheet velocities. In Greenland, the form of the subglacial hydrologic system encountered by increasing surface melt water will determine the influence of changing atmospheric conditions on ice-sheet mass balance. In Antarctica, subglacial lakes have the capacity to both modulate velocities in ice streams and outlet glaciers and provide nucleation points for new fast ice-flow tributaries. Climate models of ice-sheet responses to global change remain incomplete without a parameterization of subglacial hydrodynamics and ice dynamics.

Sulphur isotope data from early Archaean rocks suggest that microbes with metabolisms based on sulphur existed almost 3.5 billion years ago, leading to suggestions that the earliest microbial ecosystems were sulphur-based. However, morphological evidence for these sulphur-metabolizing bacteria has been elusive. Here we report the presence of microstructures from the 3.4-billion-year-old Strelley Pool Formation in Western Australia that are associated with micrometre-sized pyrite crystals. The microstructures we identify exhibit indicators of biological affinity, including hollow cell lumens, carbonaceous cell walls enriched in nitrogen, taphonomic degradation, organization into chains and clusters, and delta13C values of -33 to -460/00 Vienna PeeDee Belemnite (VPDB). We therefore identify them as microfossils of spheroidal and ellipsoidal cells and tubular sheaths demonstrating the organization of multiple cells. The associated pyrite crystals have Delta33S values between -1.65 and +1.430/00 and delta34S values ranging from -12 to +60/00 Vienna Canyon Diablo Troilite (VCDT). We interpret the pyrite crystals as the metabolic by-products of these cells, which would have employed sulphate-reduction and sulphur-disproportionation pathways. These microfossils are about 200 million years older than previously described microfossils from Palaeoarchaean siliciclastic environments.

Historical accounts describe an earthquake and tsunami on 21 July AD 365 that destroyed cities and drowned thousands of people in coastal regions from the Nile Delta to modern-day Dubrovnik. The location and tectonic setting of this earthquake have been uncertain until now. Here, we present evidence from radiocarbon data and field observations that western Crete was lifted above sea level, by up to 10 m, synchronously with the AD 365 earthquake. The distribution of uplift, combined with observations of present-day seismicity, suggest that this earthquake occurred not on the subduction interface beneath Crete, but on a fault dipping at about 30° within the overriding plate. Calculations of tsunami propagation show that the uplift of the sea floor associated with such an earthquake would have generated a damaging tsunami through much of the eastern Mediterranean. Measurement of the present rate of crustal shortening near Crete yields an estimate of 5,000 yr for the repeat time of tsunamigenic events on this single fault in western Crete, but if the same process takes place along the entire Hellenic subduction zone, such events may occur approximately once every 800 yr.

Wetlands are vegetated regions that are inundated with water on a permanent, seasonal or intermittent basis. These ecosystems play an important role in the carbon cycle: wetlands take up and store carbon, and release carbon dioxide and methane through the decomposition of organic matter. More than 50% of wetlands are located in the high northern latitudes, where permafrost also prevails and exerts a strong control on wetland hydrology. Permafrost degradation is linked to changes in Arctic lakes: between 1973 and 2004 the abundance of lakes increased in continuous permafrost zones, but decreased in other zones. Here, we use a global climate model to examine the influence of permafrost thaw on the prevalence of high-latitude northern wetlands, under four emissions scenarios. We show that as permafrost degrades, the areal extent of wetlands declines; we found a net loss in wetland extent in the three highest emissions scenarios. We also note an initial increase in the number of days of the year conducive to wetland formation, owing to an increase in unfrozen surface moisture resulting from a lengthening of the thaw season. This is followed by a dramatic decline in the number of wetland-conducive days, owing to a deepening of the permafrost surface, and drainage of near-surface moisture to deeper soil layers. We suggest that a reduction in the areal extent and duration of wetlands will influence high-latitude carbon emissions.

The Earth's climate is changing rapidly as a result of anthropogenic carbon emissions, and damaging impacts are expected to increase with warming. To prevent these and limit long-term global surface warming to, for example, 2 °C, a level of stabilization or of peak atmospheric CO2 concentrations needs to be set. Climate sensitivity, the global equilibrium surface warming after a doubling of atmospheric CO2 concentration, can help with the translation of atmospheric CO2 levels to warming. Various observations favour a climate sensitivity value of about 3 °C, with a likely range of about 2–4.5 °C. However, the physics of the response and uncertainties in forcing lead to fundamental difficulties in ruling out higher values. The quest to determine climate sensitivity has now been going on for decades, with disturbingly little progress in narrowing the large uncertainty range. However, in the process, fascinating new insights into the climate system and into policy aspects regarding mitigation have been gained. The well-constrained lower limit of climate sensitivity and the transient rate of warming already provide useful information for policy makers. But the upper limit of climate sensitivity will be more difficult to quantify.

Rising temperatures are predicted to accelerate the decomposition of labile soil organic compounds such as proteins and carbohydrates, whereas biochemically resistant compounds, such as lipids from leaf cuticles and roots and lignin from woody tissues, are expected to remain stable on decadal to centennial timescales. However, the extent to which soil warming changes the molecular composition of soil organic matter is poorly understood. Here we examine the impact of soil warming in a mixed temperate forest on the molecular make-up of soil organic matter. We show that the abundance of leaf-cuticle-derived compounds is increased following 14 months of soil warming; we confirm this with nuclear magnetic resonance spectra of soil organic matter extracts. In contrast, we find that the abundance of lignin-derived compounds is decreased after the same treatment, while soil fungi, the primary decomposers of lignin in soil, increase in abundance. We conclude that future warming could alter the composition of soil organic matter at the molecular level, accelerating lignin degradation and increasing leaf-cuticle-derived carbon sequestration. With annual litterfall predicted to increase in the world's major forests with a 3C warming, we suggest that future warming may enhance the sequestration of cuticular carbon in soil.

Abrupt changes in the African monsoon can have pronounced socioeconomic impacts on many West African countries. Evidence for both prolonged humid periods and monsoon failures have been identified throughout the late Pleistocene and early Holocene epochs 1,2. In particular, drought conditions in West Africa have occurred during periods of reduced North Atlantic thermohaline circulation, such as the Younger Dryas cold event 1. Here, we use an ocean–atmosphere general circulation model to examine the link between oceanographic changes in the North Atlantic Ocean and changes in the strength of the African monsoon. Our simulations show that when North Atlantic thermohaline circulation is substantially weakened, the flow of the subsurface North Brazil Current reverses. This leads to decreased upper tropical ocean stratification and warmer sea surface temperatures in the equatorial South Atlantic Ocean

Abrupt periods of global warming between 57 and 50 million years ago--known as the Early Palaeogene hyperthermal events--were associated with the repeated injection of massive amounts of carbon into the atmosphere. The release of methane from the sea floor following the dissociation of gas hydrates is often invoked as a source. However, seafloor temperatures before the events were at least 4-7°C higher than today, which would have limited the area of sea floor suitable for hosting gas hydrates. Palaeogene gas hydrate reservoirs may therefore not have been sufficient to provide a significant fraction of the carbon released. Here we use numerical simulations of gas hydrate accumulation at Palaeogene seafloor temperatures to show that near-present-day values of gas hydrates could have been hosted in the Palaeogene. Our simulations show that warmer temperatures during the Palaeogene would have enhanced the amount of organic carbon reaching the sea floor as well as the rate of methanogenesis. We find that under plausible temperature and pressure conditions, the abundance of gas hydrates would be similar or higher in the Palaeogene than at present. We conclude that methane hydrates could have been an important source of carbon during the Palaeogene hyperthermal events.

The Greenland ice sheet contains enough water to raise sea levels by 7 m. However, its present mass balance and future contribution to sea level rise is poorly understood. Accelerated mass loss has been observed near the ice sheet margin, partly as a result of faster ice motion. Surface melt waters can reach the base of the ice sheet and enhance basal ice motion. However, the response of ice motion to seasonal variations in meltwater supply is poorly constrained both in space and time. Here we present ice motion data obtained with global positioning system receivers located along a 35 km transect at the western margin of the Greenland ice sheet throughout a summer melt season. Our measurements reveal substantial increases in ice velocity during summer, up to 220% above winter background values. These speed-up events migrate up the glacier over the course of the summer. The relationship between melt and ice motion varies both at each site throughout the melt season and between sites. We suggest that these patterns can be explained by the seasonal evolution of the subglacial drainage system similar to hydraulic forcing mechanisms for ice dynamics that have been observedat smaller glaciers.

Subduction zones are often characterized by wedge-shaped sedimentary complexescalled accretionary prismsthat form when sediments are scraped off the subducting plate and added to the overriding plate. Large, landward-dipping thrust faults can cut through such a prism: these faults, known as megasplay faults, originate near the top of the subducting plate and terminate at the shallow, landward edge of the prism. Megasplay faults have been the subject of numerous geological and geophysical studies, but their initiation and evolution through time remains poorly constrained. Here we combine seismic reflection data from the Nankai accretionary wedge with geological data collected by the Integrated Ocean Drilling Program (IODP) and find that the splay fault cutting this wedge initiated 1.95 Million years (Myr) ago in the lower part of the prism as an out-of-sequence thrust (OOST). After an initial phase of high activity, the movement along the fault slowed down, but uplift and reactivation of the fault resumed about 1.55 Myr ago. The alternating periods of high and low activity along the splay fault that we document hint at episodic changes in the mechanical stability of accretionary prisms.

Formic acid exerts a significant influence on atmospheric chemistry and rainwater acidity. Satellite observations and model simulations suggest that terrestrial vegetation accounts for around 90% of the formic acid produced annually.

During the last deglaciation, warming over Antarctica was interrupted by a return to colder conditions from about 14,540 to 12,760yr ago. This period, known as the Antarctic Cold Reversal, is well documented in Antarctic ice cores, but the geographic extent of the cooling throughout the Southern Hemisphere remains unclear. Here we use 10Be surface-exposure ages from two glacial moraine sets from the Southern Alps, New Zealand, to assess whether the glacier advance was associated with the Antarctic Cold Reversal. We find that widespread glacier resurgence culminated 13,000years ago, at the peak of Antarctic cooling. Subsequent glacier retreat in the Southern Alps coincided with warming in Antarctica. We conclude that the climate deterioration associated with the Antarctic Cold Reversal extended into the southern mid-latitudes of the southwestern Pacific Ocean. We suggest that the extensive cooling was caused by northward migration of the southern Subtropical Front, and concomitant northward expansion of cold Southern Ocean waters.

Throughout most of the Himalaya, slip of the Indian Plate is restrained by friction on the interface between the plate and the overlying wedge of Himalayan rocks. Every few hundred years, this interface--or décollement--ruptures in one or more Mw >=8 earthquakes. In contrast, in the westernmost Himalaya, the Indian Plate slips aseismically beneath wide plateaux fronting the Kohistan Mountains. The plateaux are underlain by viscous décollements that are unable to sustain large earthquakes. Potwar, the widest of these plateaux is underlain by viscous salt, which currently permits it to slide at rates of about 3mmyr-1 (refs , ), much slower than its 2 Myr average. This deceleration has been attributed to recently increased friction through the loss of salt from its décollement. Here we use interferometric synthetic aperture radar and seismic data to assess movement of the Kohat Plateau--the narrowest and thickest plateau. We find that in 1992 an 80 km2 patch of the décollement ruptured in a rare Mw 6.0 earthquake, suggesting that parts of the décollement are locally grounded. We conclude that this hybrid seismic and aseismic behaviour represents an evolution of the mode of slip of the plateaux from steady creep towards increasingly widespread seismic rupture.

Aerosols alter cloud density and the radiative balance of the atmosphere. This leads to changes in cloud microphysics and atmospheric stability, which can either suppress or foster the development of clouds and precipitation. The net effect is largely unknown, but depends on meteorological conditions and aerosol properties. Here, we examine the long-term impact of aerosols on the vertical development of clouds and rainfall frequencies, using a 10-year dataset of aerosol, cloud and meteorological variables collected in the Southern Great Plains in the United States. We show that cloud-top height and thickness increase with aerosol concentration measured near the ground in mixed-phase clouds-which contain both liquid water and ice-that have a warm, low base. We attribute the effect, which is most significant in summer, to an aerosol-induced invigoration of upward winds. In contrast, we find no change in cloud-top height and precipitation with aerosol concentration in clouds with no ice or cool bases. We further show that precipitation frequency and rain rate are altered by aerosols. Rain increases with aerosol concentration in deep clouds that have a high liquid-water content, but declines in clouds that have a low liquid-water content. Simulations using a cloud-resolving model confirm these observations. Our findings provide unprecedented insights of the long-term net impacts of aerosols on clouds and precipitation.

The evolution of the northwest African hydrological balance throughout the Pleistocene epoch influenced the migration of prehistoric humans. The hydrological balance is also thought to be important to global teleconnection mechanisms during Dansgaard-Oeschger and Heinrich events. However, most high-resolution African climate records do not span the millennial-scale climate changes of the last glacial-interglacial cycle, or lack an accurate chronology. Here, we use grain-size analyses of siliciclastic marine sediments from off the coast of Mauritania to reconstruct changes in northwest African humidity over the past 120,000 years. We compare this reconstruction to simulations of palaeo-humidity from a coupled atmosphere-ocean-vegetation model. These records are in good agreement, and indicate the reoccurrence of precession-forced humid periods during the last interglacial period similar to the Holocene African Humid Period. We suggest that millennial-scale arid events are associated with a reduction of the North Atlantic meridional overturning circulation and that millennial-scale humid events are linked to a regional increase of winter rainfall over the coastal regions of northwest Africa.

Climate models predict that increasing greenhouse gas levels will invigorate the circulation in the upper atmosphere. But a close look at observations of the age of stratospheric air over 30 years reveals no acceleration in the circulation.

The timing and role of exhumation in the StElias orogen, the worlds highest coastal mountain range, has been unclear. Sampling is limited to high mountain ridges that tower over widespread ice fields that sit in deeply eroded parts of the orogen. Existing bedrock studies in the region are therefore prone to bias. Here we analyse detrital material of active sediment systems in the StElias orogen to obtain age information from the inaccessible ice-covered valley bottoms. We present 1,674 detrital zircon fission-track ages from modern rivers that drain the glaciers. We find a population of very young ages of less than 3 Myr from the Seward-Malaspina glacier systems that is sharply localized in the area of the orogens highest relief, highest seismicity and at the transition from transform to subduction tectonics. Our data provide evidence for intense localized exhumation that is driven by coupling between erosion and active tectonic rock uplift.

Over the past 50 years, retreating glaciers and ice caps contributed 0.5 mm yr-1to sea-level rise, and one third of this contribution is believed to come from ice masses bordering the Gulf of Alaska. However, these estimates of ice loss in Alaska are based on measurements of a limited number of glaciers that are extrapolated to constrain ice wastage in the many thousands of others. Uncertainties in these estimates arise, for example, from the complex pattern of decadal elevation changes at the scale of individual glaciers and mountain ranges. Here we combine a comprehensive glacier inventory with elevation changes derived from sequential digital elevation models. We find that between 1962 and 2006, Alaskan glaciers lost 41.9 ± 8.6 km 3yr-1 of water, and contributed 0.12 ± 0.02 mm yr-1 to a-level rise, 34% less than estimated earlier2,3. Reasons for our lower values include the higher spatial resolution of our glacier inventory as well as the reduction of ice thinning underneath debris and at the glacier margins, which were not resolved in earlier work. We suggest that estimates of mass loss from glaciers and ice caps in other mountain regions could be subject to similar revisions.

Anna Armstrong reviews Dirty Oil by Leslie Iwerks, Dogwoof: 2010. UK release date: 19 March 2010.

Extensive valley glaciers have repeatedly covered the inner gorges of the Swiss Alps during Quaternary glaciations. Two controversial explanations of the development of the features have been proposed. In the first, the gorges would have formed anew each time the glaciers receded, through fluvial incision of the previously glaciated surfaces. Alternatively, the valleys could be palimpsest features, carved through successive glacial-interglacial cycles. Here we use topographic data derived from LiDAR measurements to show that fluvial erosion rates of 8.5-18mmyr-1 would be required to create the current relief of Swiss gorges solely during the present interglacial period. Such high rates exceed the long-term average bedrock erosion rates of even the most tectonically active regions. This scenario would also require that previously incised valleys were erased during successive glaciations by commensurately high glacial erosion rates, a suggestion that is incompatible with available constraints of exhumation from thermochronometry. We therefore suggest that the gorges observed in the Swiss Alps are resilient to repeated glaciations. Our data are most consistent with the hypothesis that gorges are progressively incised below the elevations of glacial trough valleys through multiple glaciations.

Some of Earth's greatest relief occurs where glacial processes act on mountain topography. This dramatic landscape is thought to be an imprint of Pleistocene glaciations. However, whether the net effect of glacial erosion on mountains is to increase or decrease relief remains disputed. It has been suggested that in the European Alps, the onset of widespread glaciation since the mid-Pleistocene climate transition led to the growth of large, long-lived and strongly erosive alpine glaciers that profoundly influenced topography. Here we use 4He/3He thermochronometry and thermal-kinematic models to show that the Rhône Valley in Switzerland deepened by about 1-1.5km over the past one million years. Our results indicate that while the valley was incised and back-cut, high-altitude areas were preserved from erosion. We find an approximately two-fold increase in both local topographic relief and valley concavity, which occurred around the time of the mid-Pleistocene transition. Our results support the proposed link between the onset of efficient glacial erosion in the European Alps and the transition to longer, colder glacial periods at the middle of the Pleistocene epoch.

The dwarf planet Ceres is the largest object in the asteroid belt, and is generally thought to be a differentiated body composed primarily of silicate materials and water ice. Some remotely observed features, however, indicate that Ceres may instead have a composition more similar to that of the most common types of carbonaceous meteorite. In particular, Ceres has been shown to have a distinct infrared absorption feature centred at a wavelength of 3.06 m that is superimposed on a broader absorption from 2.8 to 3.7 m (refs5,8), which suggests the presence of OH- or H 2 O-bearing phases. The specific mineral composition of Ceres and its relationship to known meteorite mineral assemblages, however, remains uncertain. Here we show that the spectral features of Ceres can be attributed to the presence of the hydroxide brucite, magnesium carbonates and serpentines, a mineralogy consistent with the aqueous alteration of olivine-rich materials. We therefore suggest that the thermal and aqueous alteration history of Ceres is different from that recorded by carbonaceous meteorites, and that samples from Ceres are not represented in existing meteorite collections.

Precambrian banded iron formations provide an extensive archive of pivotal environmental changes and the evolution of biological processes on early Earth. The formations are characterized by bands ranging from micrometre- to metre-scale layers of alternating iron- and silica-rich minerals. However, the nature of the mechanisms of layer formation is unknown. To properly evaluate this archive, the physical, chemical and/or biological triggers for the deposition of both the iron- and silica-rich layers, and crucially their alternate banding, must be identified. Here we use laboratory experiments and geochemical modelling to study the potential for a microbial mechanism in the formation of alternating iron–silica bands. We find that the rate of biogenic iron(III) mineral formation by iron-oxidizing microbes reaches a maximum between 20 and 25 °C. Decreasing or increasing water temperatures slow microbial iron mineral formation while promoting abiotic silica precipitation. We suggest that natural fluctuations in the temperature of the ocean photic zone during the period when banded iron formations were deposited could have led to the primary layering observed in these formations by successive cycles of microbially catalysed iron(III) mineral deposition and abiotic silica precipitation.

Anna Armstrong reviews Big River Man by John Maringouin, Self Pictures: 2009. UK release date: 4 September 2009.

During the mid- and Late Cretaceous period, North America was split by the north-south oriented Western Interior Seaway. Its role in creating and maintaining Late Cretaceous global greenhouse conditions remains unclear. Different palaeoceanographic reconstructions portray diverse circulation patterns. The southward extent of relatively cool, low-salinity, low-δ18O surface waters critically distinguishes among these models, but past studies of invertebrates could not independently assess water temperature and isotopic compositions. Here we present oxygen isotopes in biophosphate from coeval marine turtle and fish fossils from western Kansas, representing the east central seaway, and from the Mississippi embayment, representing the marginal Tethys Ocean. Our analyses yield precise seawater isotopic values and geographic temperature differences during the main transition from the Coniacian to the early Campanian age (87-82 Myr), and indicate that the seaway oxygen isotope value and salinity were 2‰ and 3‰ lower, respectively, than in the marginal Tethys Ocean. We infer that the influence of northern freshwater probably reached as far south as Kansas. Our revised values imply relatively large temperature differences between the Mississippi embayment and central seaway, explain the documented regional latitudinal palaeobiogeographic zonation and support models with relatively little inflow of surface waters from the Tethys Ocean to the Western Interior Seaway.

Pronounced warming in the Arctic region, coined Arctic amplification, is an important feature of observed and modelled climate change. Arctic amplification is generally attributed to the retreat of sea-ice and snow, and the associated surface-albedo feedback, in conjunction with other processes. In addition, the predominant thermal surface inversion in winter has been suggested to pose a negative feedback to Arctic warming by enhancing infrared radiative cooling. Here we use the coupled climate model EC-Earth in idealized climate change experiments to quantify the individual contributions of the surface and the atmosphere to infrared radiative cooling. We find that the surface inversion in fact intensifies Arctic amplification, because the ability of the Arctic wintertime clear-sky atmosphere to cool to space decreases with inversion strength. Specifically, we find that the cold layers close to the surface in Arctic winter, where most of the warming takes place, hardly contribute to the infrared radiation that goes out to space. Instead, the additional radiation that is generated by the warming of these layers is directed downwards, and thus amplifies the warming. We conclude that the predominant Arctic wintertime temperature inversion damps infrared cooling of the system, and thus constitutes a positive warming feedback.

Earth's climate has oscillated between short-lived interglacial and extended glacial periods for the past million years. Before the last interglacial, absolutely dated markers of sea level become increasingly rare; hence, our knowledge of sea-level change driven by the waxing and waning of continental ice sheets before that time is largely based on proxy records from deep-sea cores(1-3) that lack direct age control. Here we present precise U-Th ages for a remarkable collection of submerged speleothems(4,5) from Italy, which record three sea-level highstands during the penultimate interglacial period, Marine Isotope Stage 7, from 245,000 to 190,000 years ago. We find that sea level rose above-18m (relative to modern sea level) several thousand years before maximum Northern Hemisphere insolation during the first and third highstands. In contrast, the second highstand, Marine Isotope Stage 7.3, is essentially synchronous with the insolation maximum, and sea level during this highstand only peaked at about 18 m, even though the concurrent insolation forcing was the strongest of the three highstands. We attribute the different phasing and amplitude of the Marine Isotope Stage 7.3 highstand to the extensive continental glaciation that preceded it. This finding highlights the significance of cryosphere response time to the climate system.

Purported 3,465-million-year-old microfossils from Australia have been the subject of considerable debate. A method to distinguish between pristine fossils, mineral artefacts and subsequent microbial contamination will aid the search for ancient biogenic material.

Sea ice is an integral component of the climate system, but a difficult one to reconstruct. Biochemical tracers preserved in marine sediments now reveal the waxing and waning of sea ice since the Last Glacial Maximum in an Arctic Ocean gateway.

The climate of early Mars could have supported a complex hydrological system and possibly a northern hemispheric ocean covering up to one-third of the planets surface. This notion has been repeatedly proposed and challenged over the past two decades, and remains one of the largest uncertainties in Mars research. Here, we used global databases of known deltaic deposits, valley networks and present-day martian topography to test for the occurrence of an ocean on early Mars. The distribution of ancient martian deltas delineates a planet-wide equipotential surface within and along the margins of the northern lowlands. We suggest that the level reconstructed from the analysis of the deltaic deposits may represent the contact of a vast ocean covering the northern hemisphere of Mars around 3.5 billion years ago. This boundary is broadly consistent with palaeoshorelines suggested by previous geomorphologic, thermophysic and topographic analyses, and with the global distribution and age of ancient valley networks. Our findings lend credence to the hypothesis that an ocean formed on early Mars as part of a global and active hydrosphere.

The mantle transition zone, located at depths of 410-660km between the lower and upper mantle, is an important water reservoir in the Earth's interior. However, there are regional-scale heterogeneities in the distribution of water. The zone beneath northeast China, in particular, is remarkably hydrous, but when and how it became hydrous remains uncertain. Here we combine analyses of the geochemistry of late Cenozoic basalts in northeast China with published geochemical analyses. We find a spatial correlation between basalt geochemistry and the distribution of a low-velocity zone in the underlying mantle that is interpreted as a plume upwelling from the mantle transition zone. We therefore use the basalt geochemistry to infer the composition of the mantle transition zone. The basalts have high Ba/Th and 207Pb/206Pb ratios, which we suggest record an ancient hydration event in the transition zone that occurred more than one billion years ago, probably as a result of dehydration of a subducted slab. We suggest that this ancient hydration event, combined with a more recent hydration event linked to dehydration of the subducted Pacific slab, can account for the hydrous nature of the mantle transition zone beneath China. Our results demonstrate that the mantle transition zone can remain as a stable water reservoir in Earth's interior for timescales of more than a billion years.

Fragments of ancient continental lithosphere, entrained in the shallow oceanic mantle, have been found in a number of locations in the Southern Hemisphere, including rare arc settings. Lavas erupted in these locations exhibit Pb isotopic characteristics that are similar to the so-called enriched mantle 1 reservoir, one of the end-members that define the isotopic composition of the Earth's mantle. However, no lavas with isotopic signatures resembling enriched mantle 1 have been identified in the Caribbean region. Here we present isotopic analyses for mafic-alkaline lavas from Quaternary volcanic centres in Hispaniola. We identify unusual isotopic characteristics indicating the presence of a mantle component similar to enriched mantle 1 beneath Hispaniola. Furthermore, we find evidence for an involvement of this mantle component in the genesis of spatially associated calk-alkaline lavas. On the basis of these isotopic systematics we estimate that the mafic-alkaline lavas are derived from an ancient lithospheric fragment with affinities to the supercontinent Gondwana. We conclude that the fragment originated from the Grenvillian terranes of Central America and Mexico, which also have affinities to Gondwana, indicating that Hispaniola interacted tectonically with these terranes.

Rapid global warming marked the boundary between the Palaeocene and Eocene periods 55.6 million years ago, but how the temperature rise was initiated remains elusive. A catastrophic release of greenhouse gases from the Kilda basin could have served as a trigger.