Abstract The weak, non-significant correlation between East Asian summer monsoon (EASM) rainfall and the concurrent ENSO index has hindered the development of ENSO-based forecasts of EASM. We show that this apparent statistical independence does not stem from randomness, but rather from two opposing phase relationships between the Pacific-Japan (PJ) teleconnection pattern and ENSO. A positive PJ can be triggered by two distinct Indo-Pacific sea surface temperature (SST) anomaly patterns corresponding to opposite ENSO phases: a La Niña–related tripolar pattern and an El Niño–related dipolar pattern. This results in rainfall over most PJ-affected regions of East Asia being statistically uncorrelated with the ENSO index. The PJ teleconnection patterns generated under these two scenarios exhibit notable discrepancies in their spatial morphology. This study explains the unexpectedly weak ENSO–EASM correlations and provides guidance for EASM rainfall prediction based on summer ENSO conditions. It also indicates that studies investigating the impacts of boreal summer-time ENSO should account for the diverse tropical SST patterns associated with ENSO rather than relying solely on a single ENSO index.

Ural blocking (UB) and the associated Warm Arctic–Cold Eurasia (WACE) pattern are typically linked to the negative Arctic Oscillation (AO). However, robust UB events are surprisingly observed even during the positive AO phase, a condition generally expected to suppress blocking due to enhanced zonal flow. This study investigates how positive AO magnitude modulates UB persistence. We find that under strong positive AO conditions (AO > +1), UB events persist significantly longer (6.1 days) than weak positive AO (4.7 days). This enhanced persistence results from organized North Atlantic storm tracks that facilitate intense heat and moisture into the Barents–Kara Sea. The resulting Arctic warming and sea ice loss trigger a thermodynamic feedback loop that weakens the meridional potential vorticity (PV) gradient, effectively anchoring the UB system. Our findings reveal that strong positive AO paradoxically promotes persistent blocking through storm-PV coupling, offering critical insights for improving sub-seasonal predictions of Eurasian winter extremes.

Abstract The interannual variability of tropical Atlantic cyclones (TCs) has been linked to ENSO and the Atlantic Meridional Mode. Recent studies have also highlighted the role of the equatorial Atlantic in influencing Cape Verde TCs. However, the stationarity of these teleconnections has not yet been thoroughly examined. In this work, running correlations between the anomalous frequency of TCs originating in the Cape Verde region and tropical sea surface temperatures (SSTs), reveal a shift in recent decades. Although the overall frequency of Atlantic TCs has increased since the 2000’s, changes in atmospheric and oceanic background conditions have weakened the influence of SSTs on Cape Verde TCs. Furthermore, our assessment of eddy kinetic energy as an indicator of African easterly wave activity suggests that, in the context of global warming, the predictive power of sea surface temperatures may be diminished.

Abstract Artificial intelligence has improved the accuracy and efficiency of weather forecasting, surpassing traditional numerical weather prediction models. However, the coarse spatial resolution of global weather forecasting systems limits their ability to capture fine-scale surface heterogeneity and localized extremes, particularly in regions with complex terrain or urban heat island effects. Here, we introduce SR-Weather, a deep learning-based super-resolution framework that converts coarse 0.25° forecasts into 1-km surface air temperature fields using MODIS-derived temperature targets and high-resolution auxiliary inputs. SR-Weather outperforms existing super-resolution methods by explicitly incorporating spatial context, such as topography, impervious surface fraction, and seasonal climatology maps of air temperature. When SR-Weather was applied to the FuXi global weather forecast, the 7-day forecast error in South Korea decreased by more than 20%, which was comparable to the 1-day forecast error from low-resolution prediction using simple spatial interpolation. In addition, SR-Weather effectively reconstructs missing pixels in MODIS-derived air temperature maps under heavy cloud contamination by leveraging auxiliary variables and climatologically smoothed fields. Although validated over South Korea, the framework relies on globally available MODIS products and minimal auxiliary inputs, making it feasible to retrain for other regions. These results indicate that SR-Weather is a scalable and high-fidelity tool for enhancing machine learning-based weather forecasts at fine spatial scales.

Sea surface temperatures (SSTs) in the tropical North Atlantic have historically served as reliable predictors of early-season rainfall across the Caribbean. In particular, rainfall onset has been linked to SSTs exceeding the convective threshold necessary to support deep convection. However, recent warming trends appear to have altered this relationship. Here, we show that although SSTs routinely exceed the convective threshold earlier in the season, early rainfall has not increased. This decoupling reflects a shift in the atmospheric state, with enhanced stability, evidenced by reduced convective available potential energy and increased convective inhibition, increasingly suppressing convection. Reduced rainfall results in a more persistent Caribbean Low-Level Jet (CLLJ), further inhibiting rainfall by promoting subsidence and dry air advection. Correlations indicate that dynamic atmospheric variables now explain a larger share of rainfall variability than absolute SSTs. These findings signal a regime shift in Caribbean rainfall dynamics and raise concerns about the declining utility of SST-based predictors under continued climate warming. These results have significant implications for seasonal forecasting and adaptation planning across Caribbean Small Island Developing States.

Extreme precipitation (EP) is a major climate risk, yet its projections remain uncertain due to the combined influence of thermodynamic (TH) and dynamic (DY) processes. Using multi-model simulations under three emission scenarios, we separate TH and DY contributions to the annual maximum 1-day precipitation (Rx1Day) and quantify their uncertainties. TH consistently intensifies extremes with warming, while DY strongly modulates their magnitude and direction. DY processes dominate Rx1Day uncertainty, with internal variability within DY emerging as the leading contributor. Signal-to-noise ratio (SNR) analysis shows that the forced signal emerges more clearly for TH than DY, where chaotic variability fundamentally limits predictability. The strongest intensification occurs in equatorial regions, raising equity concerns for vulnerable populations. These results demonstrate that DY internal variability is the primary driver of EP uncertainty, highlighting limits to long-term predictability and the importance of properly representing natural dynamical fluctuations in future projections.

Marker components are critical for accurately identifying fine particle (PM2.5) sources. Current marker identification is subjective, and the widely adopted markers to distinguish sub-types of biomass burning (BB) emissions are impossible. Here, PM2.5 samples from the combustion of ten biomass types were collected, and 102 chemical components were analyzed. Results indicated that the mass fractions of K⁺, K, and levoglucosan (LEV) varied in 0.1%–8.4% (2.4 ± 2.2%), 0.1%–14.1% (3.2 ± 3.1%), and 0.1%–4.9% (2.1 ± 1.4%). Source profiles for the ten sub-types of BB differed notably according to their coefficient of divergence values (0.33–0.72). It implied that there existed species which can separate PM2.5 from the ten types of BB. Ten widely used specific species ratios in literatures cannot effectively distinguish sub-types of BB due to the value overlap. A k-means identified LEV and fatty acid (C16:0) as the common markers for the ten sub-types of BB. C16:0 was firstly proposed and verified as a marker for BB here. With a random forest and the current dataset, the subtype-specific markers for only four sub-type sources were identified and verified. We indicated that there is no necessary relationship between markers and their mass percentages. This study provides a basic database and scientific reference for sub-type source apportionment and species-based emission inventory developing.

Future changes in extratropical cyclones (ETCs), their moisture transport, and their connection to atmospheric rivers (ARs) remain uncertain. Here, using downscaled simulations from the Community Earth System Model version 2 under the high-emissions scenario SSP5-8.5, we show that moisture uptake associated with intense North Atlantic ETCs will increase by ~14.5%, or ~3.29% per Kelvin of global warming by the end of this century. Subregional analysis revealed greater increases during the most intense cyclone stages, with changes of 52.1% (11.8% K−1) in the eastern North Atlantic, 33.4% (7.6% K−1) in the western North Atlantic, and 29.1% (6.6% K−1) in the northern North Atlantic. Rather than a consistent increase within existing climatological source regions, the moisture origin exhibits a geographic shift, likely linked with changes in storm-track pathways and ocean circulation under warming. Using a novel method to couple ETC moisture sources with ARs, we demonstrate that while this coupling continues to be a defining characteristic at the basin scale, the strength of their association is projected to weaken. Historical differences between low- and fully associated ETCs, particularly in terms of moisture sources and proximity to the AR axis, diminish and lose significance in a warming climate.

The evolution of the Asian Summer Monsoon (ASM) over geological timescale remains uncertain1–3 despite its fundamental role in shaping regional climate4,5, ecosystems6,7, and civilizations8. Using a series of time-slice simulations with a paleo-climate model, we assess how India–Eurasia collision tectonics9,10, Tibetan Plateau (TP) uplift11,12, and atmospheric CO2 variability13 influenced ASM evolution through the Cenozoic. Our simulation-based results suggest that ASM intensification was contingent on the TP exceeding an areal mean elevation of ~ 3.5 km in the late Eocene–Oligocene (27–38 million years ago, Ma), which strengthened the upper-tropospheric temperature gradient, promoted the seasonal northward shift of the Intertropical Convergence Zone (ITCZ), and restructured atmospheric circulation. Initially confined to East Asia14, monsoonal rainfall expanded across South Asia by the Oligocene, coinciding with enhanced circulation and reversing the meridional relative sea surface temperature gradient in the Neotethys. While TP uplift played the primary role in early ASM evolution, declining atmospheric CO2 levels became increasingly influential after the late Miocene. These findings, supported by sedimentary records of weathering and erosion15,16, underscore the dominant role of TP in climate–tectonic interactions and ASM evolution over geological timescales.