地球科学最新文献

Reef-scale climate projections, such as those generated by CMIP6, are critical for guiding the development of effective intervention strategies for mass coral bleaching events. We developed a machine learning (ML) model based on a super resolution deconvolutional neural network to rapidly downscale sea surface temperature (SST) on the Great Barrier Reef (GBR). When downscaling 80 km data to 10 km resolution, the ML model outperforms conventional interpolation methods by capturing the spatial variability of SST and extreme thermal events. We applied this model to independent datasets from both present-day and future climates, demonstrating its robustness. Additionally, we demonstrated the ML model's capability to reconstruct the spatial variability of degree heating weeks for coral bleaching risk analysis. With its ease of implementation and low computational cost, this ML model could be readily used or easily trained to rapidly downscale climate model outputs for coral reefs around the world.

Following Pinedale deglaciation (~12 ka), unconfined valleys in the Rocky Mountains experienced periods of fluvial aggradation and incision, creating distinctive valley morphologies and substrates that influence present-day hydrological and ecological characteristics. Valley floors are thus physically important sediment storage sites that preserve alluvial records of past landscape dynamics. Using geologic mapping, ground-penetrating radar surveys, sediment coring, and radiocarbon and optically stimulated luminescence (OSL) geochronology, we investigated an unconfined portion of the South Fork Cache la Poudre River Valley, Colorado Front Range, to identify the dominant processes and temporal patterns of valley alluviation and incision following glacial retreat. We mapped a variety of glacial and fluvial deposits in the valley including two till deposits, distinct outwash terraces, fluvial terraces and an extensive floodplain. Abundant glaciofluvial outwash (65 m observed at one drill site) was deposited up-valley of the Last Glacial Maximum terminal moraine. Lateral bar migration, channel filling and vertical accretion of sediments were important processes of outwash aggradation and floodplain deposition. OSL dating of unconsolidated, laminated sand and silt suggests ponding up-valley of the terminal moraine between 13.4 and 11.5 ka, and the potential for an outburst flood(s). Channel incision occurred prior to 7.8–1.5 ka, creating outwash terraces that comprise over 30% of the valley floor area. Sedimentation occurred on the fluvial terrace and floodplain from at least 2.1 to 1.3 ka. The modern floodplain has been aggrading for at least 500 years. The South Fork Valley has anomalously thick post-glacial sediment from lateral migration and channel filling, whereas other Colorado headwater valleys are dominated by mass wasting deposits, beaver pond sediments or fluvial vertical accretion. The relict glacial topography and low hillslope-floodplain connectivity exerts the strongest control on alluvial dynamics in the South Fork Valley. Results of this study broaden the foundation for understanding post-glacial alluvial dynamics in unconfined mountain valleys. Knowledge of the processes that create and maintain alluvial fills is critical for effective management of these valleys.

The magnetotail current sheet of Mars exhibits a dawn-to-dusk asymmetry that has been seen in satellite observations and MHD simulations. However, the influence of season has not been thoroughly investigated in MHD simulations. This investigation incorporates seasonal variations, driven by planetary eccentricity and solar variability, within the BATS-R-US multispecies MHD code to examine the influence of crustal magnetic fields and the ionosphere on the magnetotail current sheet. The solar wind interaction at Mars is analyzed for the following cases: solar maximum at perihelion (PERMAX), solar maximum at aphelion (APHMAX), solar minimum at perihelion (PERMIN), and solar minimum at aphelion (APHMIN). Simulation results show that the current sheet exhibits a duskward shift at solar maximum and a dawnward shift at solar minimum. In simulations that omit the crustal sources, the current sheet remains symmetric along the $��Y�=�0�$ plane. Because these results did not induce a shift, the ionization rates were adjusted for the PERMAX and APHMIN cases. The ionization rates were increased by four orders of magnitude in the PERMAX case, but the current sheet remained symmetric. However, the current sheet in the APHMIN case shifted slightly duskward when the ionization rates were decreased by nine orders of magnitude. It was determined that the crustal magnetic fields dominate the magnetotail current sheet shift, and the code setups from this investigation should be scrutinized for refined model comparison.

Earth’s atmosphere underwent permanent oxidation during the Great Oxidation Event approximately 2.45–2.22 billion years ago (Ga) due to excess oxygen (O₂) generated by marine cyanobacteria. However, understanding the timing and tempo of seawater oxygenation before the Great Oxidation Event has been hindered by the absence of sensitive tracers. Nitrogen (N) isotopes can be an indicator of marine oxygenation. Here we present an ~200 Myr nitrogen isotope oscillation recorded by Neoarchaean and Palaeoproterozoic banded iron formations from the Hamersley Basin, Western Australia, that were deposited in relatively deep marine shelf environments. Paired with the Jeerinah Formation shale record, our data from the Marra Mamba Iron Formation suggest that oxic conditions expanded to banded iron formation depositional environments from ~2.63 to 2.60 Ga. Subsequently, a positive δ¹⁵N excursion occurred in the ~2.48 Ga Dale Gorge Member, marking a decline in seawater O₂ and enhanced denitrification. This O₂ deficit was followed by a second phase of increasing O₂ levels as indicated by a gradual return to moderately positive δ¹⁵N values in the ~2.46 Ga Joffre Member and 2.45 Ga Weeli Wolli Iron Formation. These variations underscore a nonlinear history of marine oxygenation and reveal a previously unrecognized oscillation in seawater O₂ levels preceding the Great Oxidation Event.

Atmospheric rivers (ARs) in winter can induce significant melting of sea ice as they approach the ice cover. However, due to the complex physical properties of sea ice, the specific processes within the ice pack that are responsible for its response to ARs remain poorly understood. This study aims to shed light on this question using a stand-alone sea ice model forced by observed atmospheric boundary conditions. The findings reveal that the AR induced ice melt and hindered ice growth in the marginal seas are attributed to a combination of thermodynamic and dynamic processes. The AR-wind transports ice floes from the marginal seas back to the central Arctic dynamically, resulting in a thickening of the ice cover in that region. Among the thermodynamic processes, reduced congelation growth (54%–56%), enhanced basal melting (17%–26%), and inhibited snow-ice formation (11%–21%) play major roles in the sea ice loss in the marginal seas.

Sporadic E (Es) layer plays a prominent role in revealing both upward and downward atmosphere-ionosphere coupling process. This study investigates the responses of Es layers to the May 2024 super geomagnetic storm by using 37 ground-based ionosondes distributed globally and space-based COSMIC-2 radio occultation observations. The results show that Es layers were significantly enhanced during the recovery phase of geomagnetic storm. In addition, the enhanced Es layers mainly occurred over Southeast Asia, Australia, the South Pacific and the East Pacific. The temporal evolution of foEs disturbances over the Asian-Australian sector clearly shows the “wave propagation” characteristics from high to low latitudes, indicating that the enhancements of the Es layers are most likely caused by the disturbed neutral winds in the E region. This study presents observational evidence for the downward impacts of the geomagnetic storm on the E-region ionosphere.