Atmospheric and Oceanic Sciences

Abstract. Accurately characterizing aerosol vertical distributions is essential for evaluating radiative forcing and air quality. While Chemical Transport Models (CTMs) simulate spatially continuous Aerosol Extinction Coefficient (AEC, km−1), they exhibit systematic AEC biases. Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) observations provide precise AEC profiles but are constrained by sparse spatial sampling. To bridge this gap, we propose a physics-informed Transformer framework as a supervised bias-correction model to correct biases in the AEC profiles simulated by GEOS-Chem. Unlike a standard Transformer, our framework features a dual-stream architecture with explicit physical constraints. It employs gated feature fusion to integrate vertical structures (combining GEOS-Chem priors with MERRA-2 profiles) by dynamically identifying height-dependent drivers, and leverages cross-attention to incorporate MERRA-2 surface environmental constraints for modulating AEC vertical rectification with synoptic contexts. This approach effectively predicts systematic biases relative to CALIOP satellite observations and resolves AEC profiles, surpassing methods retrieving only aerosol layer heights. Leave-One-Year-Out validation over East Asia during 2017–2019 demonstrates significant AEC precision improvements, increasing R from 0.49–0.53 in the GEOS-Chem simulations to 0.66–0.73 and reducing RMSE by approximately 25 %. The model effectively mitigates over-diffusion, significantly reducing AEC simulation biases in the critical near-surface layer while capturing smoothed biomass burning and dust plumes. Additionally, it exhibits robust cross-continental transferability, reproducing bias patterns over the North American domain (R=0.70) without retraining, confirming the internalization of universal physicochemical relationships linking atmospheric states to simulation biases. Furthermore, interpretability analysis serves as a diagnostic tool to guide physical model improvement. The model identifies temperature and sensible heat flux as primary drivers to constrain boundary layer mixing, pointing to potential uncertainties in vertical eddy diffusion. Additionally, it uses environmental proxies (e.g., vegetation indices and diffuse radiation) to diagnose potential deficiencies in dust threshold friction velocity and secondary organic aerosol yields. These insights provide a physical basis for refining parameterization schemes in CTMs.

Accelerating permafrost thaw may release vast deep (>3 meters) frozen soil carbon as carbon dioxide (CO 2 ), but this magnitude remains uncertain because current Earth system models (ESMs) lack deep carbon processes. Using an updated ORCHIDEE-MICT model simulating Pleistocene Yedoma formation and Holocene peatland development, we project northern (>30°N) carbon responses under climate change. Compared to the original model, including these deep carbon pools improves agreement with observations and reduces net CO 2 uptake by 47 to 74 petagrams of carbon from 1900 to 2100 across three future scenarios because of deep carbon decomposition with accelerated active-layer deepening. Under high-emission pathways, the northern soil carbon balance shifts from a sink to a source of 32 petagrams of carbon, advancing the reversal reported in earlier studies into the 21st century. Consistent with field data, our model shows that colder soils retain more labile carbon—contrary to assumptions in many Coupled Model Intercomparison Project (CMIP) models—helping explain their persistent sink bias. Our results highlight the need to represent both the quantity and quality of permafrost carbon in ESMs.

Abstract We introduce the ZJU–SCATT–V1.0 database, a comprehensive compilation of single–scattering properties for non–spherical and inhomogeneous aerosol and ice particles. This database addresses a critical community need for improved representation of atmospheric particles, particularly aerosols and ice crystals, in remote sensing algorithms, radiative transfer models, data assimilation systems, and climate projections. ZJU–SCATT–V1.0 transcends traditional oversimplified approximations by employing advanced computational electromagnetic methods, such as the invariant imbedding T–matrix method, and sophisticated particle models based on super–spheroidal geometries. It encompasses major aerosol types including dust, black carbon, and sea salt, as well as ice crystals, covering a broad spectral range of 0.3–20 μm (extending to 100 μm for ice crystals). All models have been rigorously validated against laboratory measurements and/or remote sensing observations, establishing ZJU–SCATT–V1.0 as a vital bridge between particle microphysics and macroscopic radiative impacts. To maximize accessibility and utility, the database as well as its neural network emulator are integrated into the Deep–time Digital Earth Program platform. We welcome community feedback to evaluate and improve the ZJU–SCATT–V1.0 for diverse applications.

Artificial intelligence has advanced global weather forecasting, outperforming traditional numerical models in both accuracy and computational efficiency. Nevertheless, extending predictions beyond subseasonal timescales requires the development of deep learning (DL)-based ocean-atmosphere coupled models that can realistically simulate complex oceanic responses to atmospheric forcing. This study presents KIST-Ocean, a DL-based global three-dimensional ocean general circulation model. Comprehensive evaluations demonstrate the model's robust ocean simulation skill and efficiency. Moreover, it reproduces ocean responses, such as Kelvin and Rossby wave propagation, and vertical motions induced by wind stress curl, demonstrating its ability to represent key atmospherically forced ocean dynamics underlying climate phenomena, including the El Niño-Southern Oscillation. These findings reinforce confidence in DL-based global weather and climate models by demonstrating their capacity to capture essential ocean-atmosphere relationships. Building on this foundation, the present study paves the way for extending DL-based modeling frameworks toward integrated Earth system simulations, thereby offering substantial potential for advancing long-range climate prediction capabilities.

. Here we identify a climate-driven mechanism linking accelerating glacier disintegration to an increase in deep-sea hard-bottom habitats far beyond calving fronts. Seafloor observations in Fram Strait show a localized increase in the density and patchiness of dropstones delivered by debris-laden icebergs. At the same time, four decades of shipboard records show that the occurrence of icebergs increased abruptly in the early 2000s. Backtracking links these icebergs to the main outlet glaciers in northeast Greenland and the Russian High Arctic. In northeast Greenland, the timing of glacier destabilization coincides with this rise, whereas sparse satellite coverage in the Russian sector limits temporal attribution despite indications of enhanced glacier activity. A model sensitivity study shows that, apart from intensified calving, a more dynamic sea ice cover enhances downstream transport of glacial ice. Along these pathways, increased iceberg activity could reshape deep-sea habitats through enhanced melt and associated lithogenic input, and elevate navigational hazards as maritime traffic expands in the Arctic. Although modest compared with the iceberg discharges of Pleistocene Heinrich events, this mechanism provides a modern analogue of long-range cryospheric influence on the seafloor in a warming climate.

Abstract With global warming, the intensity and frequency of floods have markedly increased, which results in substantial losses of life and property. The Pearl River Basin (PRB) in South China, with its complex topography, remains highly susceptible to flooding. To enhance the precision of flood simulation and forecast in the PRB, an overland flow scheme was first integrated into the community Noah land surface model with multiparameterization options (Noah-MP) and subsequently coupled with the Weather Research and Forecasting (WRF) model. These models were applied to a record precipitation event occurring over the PRB in April 2024 to validate their improvements. Results reveal that the modified Noah-MP can effectively simulate hydrological processes. The cumulative surface runoff is strongly affected by topography and has a higher magnitude in low-lying areas. The accumulated water depth generally aligns with the satellite-observed inundation, and the error in soil moisture between model and observation reduces. Further, the modified WRF model has successfully reproduced the inundation area in most regions, contrasting with the original scheme's inability to simulate flooding. In addition, the improvement in hydrological processes in the modified WRF also enhances the ability to simulate precipitation through land-atmosphere interactions. A comparison with the WRF-Hydro simulations further demonstrates that our scheme achieves a certain degree of improvement in simulating inundation. This study presents a promising approach for improving flood simulations in complex topography, which is instrumental in mitigating the loss of life and property caused by flood disasters in the PRB.

Abstract As the Earth continues to warm, humid heat extremes (HHEs) have emerged as a widely recognised threat to human health in equatorial Southeast Asia (SEA). While most studies have focused on climate trends, this study presents a comprehensive analysis of the synoptic and large-scale drivers of HHEs. On daily timescales, both the Madden-Julian Oscillation (MJO) and Kelvin waves are the leading modes of HHE variability. HHE risk increases by 1.2–1.4x during the dry-to-wet transition of the MJO phases, predominantly driven by increased near-surface specific humidity preceding the peak rainfall anomaly in phase 2 and increased shortwave radiation due to reduced cloud cover in phase 8. HHE risk increases by 1.3–2.0x during the dry phase of equatorial Kelvin waves, which drives subsidence and increased shortwave warming. On interannual timescales, El Niño is the leading driver, under which HHE risk increases 3 – 5x. Despite the limited overlap (19%) between wet- and dry-bulb temperature extremes, the differences in their temperature and humidity conditions, and their drivers, are small. This understanding lays the groundwork for short-range to seasonal forecasts, which are a crucial component of much-needed heat early warning systems.