Abstract:Solar photovoltaic (PV) power generation represents one of the most-competitive and scalable options for low-carbon energy production. As a major energy consumer and carbon emitter, China's large-scale deployment of solar PV is central to achieving its energy transition and carbon neutrality goals. However, PV generation is highly sensitive to meteorological conditions, which are in turn influenced by future changes in anthropogenic emissions under carbon neutrality targets. These interactions introduce significant uncertainty for the long-term planning and optimization of solar PV deployment in China. In this study, we use the Community Earth System Model version 2 (CESM2) to conduct three sets of simulation experiments, combined with a multivariate bias correction algorithm (MBCn) to eliminate systematic model biases. We quantitatively assess the impacts of global anthropogenic carbon dioxide and aerosol emissions reduction, under a carbon neutrality scenario, on changes in PV potential (PV_POT) and the frequency of extremely low PV output events across China during the mid-21st century (2041—2060).
Results show that global anthropogenic emissions reductions under a carbon neutrality pathway significantly increase PV_POT over China by approximately 4.2%, while reducing the maximum duration (PV10D), fraction of days (PV10), and number of events (PV10N) of extremely low PV output by about 21.1%, 35.3%, and 56.2%, respectively, relative to the Shared Socioeconomic Pathway (SSP) 2-4.5, which represents a moderate emission scenario. These findings suggest that global emissions mitigation not only enhances total PV generation potential but also improves the stability of PV output. The reduction of anthropogenic emissions within mainland China accounts for roughly 77%—93% of the total increase in regional mean PV_POT and 62%—92% of the total decrease in extremely low PV output across Northeast China, Central and Eastern China, and Northwest China. In contrast, emissions reductions outside mainland China contribute more substantially to improvements over the Tibetan Plateau, explaining approximately 66% of the total increase in PV_POT and 72%—82% of the reduction in extremely low PV output. These results highlight that achieving the full potential of PV generation in the energy transition requires coordinated global emission reduction efforts. Mechanistically, the observed changes in PV_POT and extremely low PV output are primarily driven by variations in surface downwelling solar radiation, with comparatively smaller contributions from surface air temperature and wind speed.
This study provides new quantitative insights to inform effective PV planning and deployment strategies in China under carbon neutrality. Nevertheless, the findings are subject to certain limitations, as they are based on simulations from a single earth system model. Future studies incorporating multi-model ensembles and evaluating additional renewable energy sources—such as wind energy and hydropower—will be essential for developing a more comprehensive understanding of climate impacts on renewable energy systems under global decarbonization pathways.

Abstract:Inter-basin interactions between the tropical Atlantic and tropical Pacific Oceans play a pivotal role in shaping global climate variability on interannual timescales.Understanding the dynamical processes underlying these linkages is essential for improving seasonal climate prediction.The Atlantic Niño,the dominant mode of tropical Atlantic variability,exhibits two distinct forms:the Canonical Atlantic Niño (CA) and the Non-canonical Atlantic Niño (NCA).While the CA is well known to arise from the Bjerknes feedback,the mechanisms driving the NCA and its connections to preceding ENSO phases remain less clearly understood.This study systematically investigates the responses of CA and NCA to the preceding winter ENSO and their subsequent feedbacks during the ENSO decay phase,with the goals of (1) identifying the distinct physical processes through which La Niña and El Niño events trigger CA and NCA,respectively;(2) quantifying the feedback of CA and NCA on the decaying ENSO;and (3) evaluating the ability of CMIP6 climate models to reproduce these observed mechanisms.Using observational and reanalysis data spanning 1940—2022 (ERA5 and SODA),interannual variability (periods ≤ 9 years) was extracted through Fourier high-pass filtering.Composite,correlation,and regression analyses,along with Student’s t-tests,were used to examine associated oceanic and atmospheric anomalies.La Niña-CA and El Niño-NCA events were identified based on the relationship between the winter (preceding December—February),Niño3.4 index and the summer (June—August) Atlantic Niño3 index (ATL3),with events selected when either index exceeded ±0.5 standard deviations.
The results show that a preceding winter La Niña induces westerly wind anomalies over the equatorial Atlantic via the Matsuno-Gill response,triggering downwelling Kelvin waves that suppress climatological upwelling and warm the eastern equatorial Atlantic.This warming reinforces low-level westerlies through the local Walker circulation,establishing a Bjerknes feedback that develops into a CA by early summer.In contrast,a preceding winter El Niño generates anticyclonic anomalies and warm SSTs in the tropical North Atlantic through the Pacific-North American teleconnection (PNA).The warm water is subducted and transported equatorward via mean meridional currents,leading to subsurface warming and the subsequent development of an NCA.
Regarding feedback,the NCA significantly accelerates El Niño and promotes a phase transition to La Niña,whereas the CA exerts little influence on the decay of La Niña.During El Niño-NCA events,the Niño3.4 index decreases by 0.84 ℃ from April to August—significantly faster than during La Niña-CA or El Niño events.This rapid decay occurs through two pathways:1) warm SST anomalies in the tropical Indian Ocean and NCA generate eastward-propagating atmospheric Kelvin waves that produce easterly wind anomalies,moisture convergence,and enhanced convection over the Maritime Continent,thereby activating Bjerknes feedback in the Pacific;and 2) the Walker circulation response to the NCA enhances subsidence over the central-eastern Pacific,further favoring La Niña onset.
The findings demonstrate that CA and NCA are triggered by distinct mechanisms linked to preceding ENSO phases and that the NCA plays a key role in facilitating the transition from El Niño to La Niña.Multi-model ensemble mean (MME) simulations from 48 CMIP6 models under PiControl experiments generally reproduce the observed response and feedback processes,supporting the robustness of these mechanisms.Nonetheless,model biases persist,including underestimated ENSO decay rates and inter-model uncertainties in feedback strength.This study highlights the importance of accurately representing tropical inter-basin interactions in climate models to enhance the predictability of ENSO and its global climate impacts,emphasizing the need to distinguish between CA and NCA in both observations and modeling efforts.

Abstract:The Madden-Julian Oscillation (MJO), the dominant mode of intraseasonal variability in the tropical atmosphere, is characterized by large-scale deep convection coupled with planetary-scale circulation. MJO events generally originate over the Indian Ocean and propagate eastward into the western Pacific. While the convective component weakens near the dateline, the circulation component continues eastward, completing a cycle within 30—90 days. Because of its strong influence on global weather and climate anomalies, understanding MJO variability is essential for improving extended-range forecasts.
MJO activity exhibits interdecadal variability in intensity, propagation speed, and spatial extent. Previous studies have primarily focused on intensity and propagation, while the period characteristics of MJO circulation and their interdecadal variations remain less well documented. Variability in MJO period length are is particularly relevant, as prolonged periods are associated with climate extremes such as southern China heatwaves, Meiyu rainfall in the Yangtze River basin, and Coastal Niño events.
The Atlantic Multidecadal Oscillation (AMO) is a leading mode of interdecadal climate variability that modulates ENSO, western Pacific tropical cyclones, and the Asian monsoon. Recent evidence suggests that AMO may influence MJO convective propagation by altering background winds and low-level moisture over the tropical Pacific. However, whether AMO also regulates the periodic characteristics of MJO circulation has not been systematically examined.
This study investigates the role of AMO in modulating the interdecadal variability of MJO circulation periods using ERA-20C reanalysis and ERSSTv5 sea surface temperature data. The analysis addresses three questions: 1) whether AMO induces interdecadal variations in MJO circulation period characteristics, 2) the physical processes responsible for these variations, and 3) the mechanism through which AMO exerts its influence.
The results show that AMO significantly modulates MJO circulation periods. During positive AMO phases, the mean circulation period shortens to approximately 45 days, whereas during negative phases it lengthens to about 60 days, with a correlation coefficient of -0.77. This variability is mainly associated with changes in propagation speed over the Western Hemisphere: accelerated propagation during positive phases results in shorter circulation periods, while decelerated propagation during negative phases produces longer periods. The modulation of propagation speed is linked to changes in the zonal sea surface temperature gradient between the tropical central-eastern Pacific and the tropical Atlantic. Positive AMO phases feature cold anomalies in the tropical central-eastern Pacific and warm anomalies in the tropical Atlantic, which enhance the zonal gradient and favor faster propagation.
Although the circulation period varies with AMO phase, the convective period remains stable, with power spectra consistently peaking near 60 days within the 30—90-day band. This indicates that the coupling between MJO convection and circulation is modulated on interdecadal timescales.
These findings demonstrate a trans-basin linkage between AMO and the interdecadal variability of MJO circulation periodicity, highlighting the role of AMO in regulating intraseasonal variability through large-scale ocean-atmosphere interactions.

Abstract:In recent decades, inter-monthly variability of winter surface air temperature (SAT) in East Asia has become increasingly pronounced, often manifesting as sharp transitions between extreme cold and extreme warm events across different months or within different stages of winter. In China, SAT exhibits distinct month-to-month fluctuations, with some winters beginning anomalously warm and ending unusually cold, or vice versa. Previous studies have suggested that these fluctuations are influenced by atmospheric circulation anomalies and modulated by extratropical teleconnections in the mid-high latitudes. However, the specific teleconnection that exert dominant influence, as well as their relative roles, remain unclear.
This study investigates the spatial and temporal characteristics of inter-monthly SAT anomalies in China and investigates the impacts of Eurasian mid-high latitude teleconnection patterns on SAT variability during December-January-February (DJF) over the period 1979—2020. The leading empirical orthogonal function mode (EOF1) of monthly SAT anomalies reveals a pan-China variability pattern, characterized by two distinct types: 1) uniform anomalies of the same sign across DJF, and 2) anomalies of opposite sign between early winter (December) and late winter (next January-February). Together, these two types account for approximately 50% of the total EOF1 variance.
Analysis shows that the two types are associated with markedly different atmospheric circulation anomalies. Opposite-sign anomalies between early and late winter are primarily linked to the Polar-Eurasia (POL), East Atlantic (EA), and Eurasian (EU) patterns, while uniform anomalies across DJF are more strongly associated with the Arctic Oscillation (AO), POL, Scandinavian (SCA), western Pacific (WP), and EU patterns. These teleconnection anomalies alter the location and intensity of the Siberian high, thereby exerting significant control over inter-monthly SAT variability.
The results highlight the dominant role of anomalous teleconnection patterns on monthly timescales in shaping the leading mode (EOF1) of winter SAT variability in China. This study provides new insights into the mechanisms driving inter-monthly temperature fluctuations, with implications for seasonal climate prediction.

Abstract:Global warming and associated climate change have intensified the frequency and severity of extreme summer events across the Northern Hemisphere, posing significant risks to human health and socioeconomic systems. Increasingly, these events exhibit strong spatial connectivity, manifested as the concurrent concurrence of heatwaves, droughts, floods, and wildfires in distant regions. Understanding the interactions and forcing mechanisms behind such compound events is critical for improving prediction skill and informing climate risk management. This study investigates whether the simultaneous extreme heat events in western Europe and the Yangtze River basin in July 2022 were linked to the unprecedented warming over the Qinghai-Xizang Plateau during that summer. Using observational datasets, NCEP/NCAR analysis, and CPC precipitation data, we analyze the evolution of heatwave characteristics and associated circulation anomalies, with emphasis on the physical processes through which anomalous plateau warming modulated Rossby wave propagation. Results show that from 2 to 13 July 2022, the Qinghai-Xizang Plateau experienced extreme warming. Subsequently, from 14 to 25 July, the Yangtze River basin suffered its most severe heatwave-drought event since the founding of the People's Republic of China, while western Europe was simultaneously impacted by intense heatwaves. The analysis reveals that plateau warming induced a northward shift and strengthening of the southern branch of the midlatitude westerly jet. Superimposed upstream wave activity flux reinforced mid-high-latitude circulation anomalies, including an intensified and eastward-extending European ridge, a strengthened South Asian high, and an anomalous western Pacific subtropical high over eastern China. These anomalies jointly established an atmospheric teleconnection linking the two heat extremes. The findings highlight that extreme Qinghai-Xizang Plateau warming initiated a Rossby wave train that generated anticyclonic circulation anomalies over both Europe and the Yangtze River basin, leading to their concurrent extreme heat events. Future studies could develop an index of plateau warming to test its statistical relationship with historical concurrent heat events across Eurasia. Moreover, the Yangtze River basin event in 2022 likely reflected the combined influence of multiple external forces, whose synergistic effects were not addressed here but warrant further investigation.

Abstract:The southeastern Qinghai-Xizang Plateau is characterized by frequent convective activity. The Yarlung Tsangpo Grand Canyon (YTGC) serves as a key water vapor channel for the plateau, making it crucial to understand the evolution of convective systems in this region for improving plateau weather and climate prediction. Previous studies have identified two distinct convective activity zones near 30°N on the plateau; however, convective frequency is notably low directly over the YTGC. Convective systems originating over the western plateau are generally confined west of 95°E, with few propagating eastward. The reason for this frequent interruption of convective activity over the YTGC remains unclear. To address this issue, CLDAS-v2.0 precipitation data were used to analyze 44 summer precipitation events from 1998 to 2019. Based on propagation characteristics, the events were classified into two types: “continuous” and “interrupted”. The classification revealed a precipitation minimum center over the YTGC, where the eastward propagation of most convective systems is blocked. Only a small subset of systems successfully propagate eastward through the canyon.
Because reanalysis data lack sufficient resolution to resolve convective processes under the complex terrain of the plateau's southern slope, high-resolution numerical simulations were conducted. The simulations reproduced both types of precipitation events. Diagnostic analysis showed that the key difference at 500 hPa between the two types lies in the location of convergence: in the “continuous type”, the convergence center is located directly over the YTGC, whereas in the “interrupted type” it shifts to higher latitudes. This difference is associated with stronger northerly winds in the “continuous type”. Positive pressure kinetic energy eqution diagnosis further indicated that anomalous northerlies in the “continuous type” enhance the meridional wind convergence term, increasing perturbation kinetic energy and favoring the maintenance and development of convective systems. From a non-equilibrium perspective, the “continuous type” is characterized by a persistent negative divergence tendency zone along the YTGC latitude band, while the “interrupted type” exhibits a transition from negative to positive values from west to east across, creating unfavorable conditions for eastward propagation. Vertical meridional circulation analysis revealed that the “continuous type” exhibits a stronger northerly flow, part of which descends along the southern slope into the canyon, converges at lower latitudes, and induces upward motion. This ascending branch transports moisture to the mid-levels, where it is carried onto the plateau by southerly winds. Coupled with favorable ascent within the YTGC itself, this process enhances precipitation. A sensitivity experiment was further performed by artificially weakening northerly winds in a “continuous” case. The modification caused precipitation over the YTGC to be interrupted, confirming that northerly winds are a crucial factor regulating precipitation processes in the canyon region.

Abstract:This study investigates the dynamics of extreme precipitation in the middle and lower Yangtze River basin (MLYRB) from 1980 to 2020. Using surface observations and ERA5 hourly reanalysis, statistical techniques, including trend analysis and the Mann-Kendall test, were applied to examine spatiotemporal variability in daily and hourly extreme precipitation. The main findings are as follows: 1) Both the amount and intensity of daily precipitation in the MLYRB show an increasing trend, while the frequency and spatial extent of heavy rainfall events have declined sharply. This indicates increasing concentration of daily extremes, thereby amplifying associated risks. 2) Extreme hourly precipitation occurs predominantly in the lower Yangtze River, although less frequently, with intensity increasing from the southwest to the northeast, a pattern likely shaped by topography. Mountainous regions experience weaker rainfall, whereas plains are subject to stronger extremes. 3) Multi-scale analyses reveal marked heterogeneity across temporal scales. Annually, the frequency of extreme hourly rainfall is decreasing, along with reduced variability, but the intensity of individual events continues to strengthen. At seasonal and monthly scales, summer presents the greatest risk, with June showing the highest dispersion and strongest variability, while winter exhibits the weakest activity. The high frequency of precipitation events in spring further elevates flood risk. 4) Peak values of extreme hourly rainfall exhibit only modest downward trends and limited interannual variability. Event frequency is concentrated in three main periods (01:00 BST—07:00 BST, 08:00 BST—19:00 BST, and 20:00 BST—24:00 BST), each showing an increasing trend. Late-stage rainfall dominates extreme hourly precipitation in the MLYRB, occurring far more frequently than other types and remaining stable over long timescales, indicating persistence in temporal distribution patterns. These results suggest that mitigation strategies for flood and waterlogging should account for the concentration and persistence of extreme hourly rainfall in the MLYRB. By analyzing precipitation dynamics across multiple scales, this study provides new insights into the mechanisms of extreme rainstorm events, offering a scientific basis for policy development, urban planning, and climate adaptation. The findings also contribute to disaster risk reduction by supporting more effective early warning systems and enhancing community resilience to extreme weather events.

Abstract:The tropospheric quasi-biennial oscillation (TBO) is a mode of climate variability with a period of approximately 2—3 years, primarily observed in tropical, subtropical, and mid- to high-latitude regions of Eurasia and the Southern Hemisphere. It manifests as quasi-periodic variations in atmospheric circulation, precipitation, sea surface temperature (SST), and snow cover. In China, a prominent quasi-2-year is evident in summer precipitation, particularly over the lower reaches of the Yangtze River Valley (LYRV), which lies within the East Asian subtropical monsoon region and exhibits pronounced TBO characteristics. Although the TBO is closely associated with large-scale climate modes such as the El Niño-Southern Oscillation (ENSO), its core driving mechanisms involve tropospheric dynamics, ocean-atmosphere interactions, and connections with stratospheric circulation. The TBO represents a critical timescale bridging annual cycles and interannual variability (e.g., ENSO). Understanding its evolution is essential for extending seasonal-to-interannual climate prediction lead times (approximately 6—18 months). The TBO is also closely linked to the variability of intraseasonal oscillations (ISO) and to extreme climate events such as monsoon precipitation anomalies, droughts, and heatwaves, thereby providing valuable guidance for agricultural planning, water resource management, and disaster mitigation.
This study develops a data-driven prediction model for interannual variations in the TBO component of rainfall. The quasi-2-year components (TBO) of monthly precipitation in the LYRV and the principal components (quasi-biennial oscillation, QBO) of the 50 hPa stratospheric zonal and meridional winds for 1979—1998 were used to construct a time-varying Extended Complex Autoregressive (ECAR) model for predicting the QBO-related component of rainfall in the LYRV. An independent 12-year real-time interannual prediction experiment (1999—2020) was conducted on the quasi-biennial component of monthly precipitation over the LYRV. The results demonstrate that the ECAR model exhibits high predictive skill, maintaining strong forecast accuracy up to a 15-month lead time—significantly outperforming the conventional autoregressive (AR) model. These forecasts provide valuable predictive guidance for anticipating summer flood processes in the LYRV more than a year in advance.
The proposed data-driven prediction method employs real-time singular spectrum analysis (RSSA) to extract the TBO components from the troposphere and QBO components from the stratosphere, both characterized by strong autocorrelation. Through Fourier transformation, these primary quasi-2-year components are converted into complex low-frequency signals in the frequency space, forming an extended complex matrix that captures evolving relationships among atmospheric variables. This new set of variables to better jointly shape a new pattern of variable changes. From the perspective of multivariate synergy, collaborative patterns that are difficult to be identified by traditional methods can be uncovered.A simplified, time-varying ECAR model is then derived to represent the dynamic interactions among these components. The inverse Fourier transform yields the predicted vectors in the original space. This framework effectively reduces data diversity, simplifies complex relationships, and adapts to interdecadal changes in coupling among low-frequency processes, thereby enhancing forecast skill and extending prediction lead times. Unlike traditional physics-based numerical models or AI (artificial intelligence) systems constrained by initial conditions and model complexity, this data-knowledge-simplification approach provides a robust alternative for interannual climate prediction. It captures real-time global QBO signals and the synergistic effects of tropical and extratropical stratospheric QBOs on tropospheric TBO-related precipitation over the lower Yangtze River region, substantially improving interannual predictability of the TBO. When combined with interdecadal trends and sub-seasonal precipitation variability, this approach enhances the predictive capability for summer rainstorms and flood events across the LYRV.

Abstract:Warm-season precipitation systems in China, influenced by monsoon circulations, diverse synoptic conditions, and complex topography, frequently generate heavy rainfall and severe convective weather, often resulting in devastating hydrometerological disasters. Traditional monitoring tools such as ground-based radars and geostationary satellites have inherent limitations in spatial coverage and in resolving internal precipitation structures. The advent of precipitation measurement satellites—TRMM (1997—2015), GPM (2014—present), and China's FY-3G (2023—present)—equipped with spaceborne precipitation radars, has revolutionized this field. These satellites provide long-term, high-quality 3D observations, enabling systematic analysis of precipitation system characteristics over nearly three decades. This review synthesizes recent research progress in China, summarizing advances in the characterization of precipitation systems using TRMM and GPM data. It focuses on identification methodologies, climatological characteristics, and the physical linkages between 3D structures and severe weather. The core approach involves objective identification using connected-component analysis of radar reflectivity. Systems are parameterized by key metrics such as area, eccentricity, convective-stratiform ratio, and vertical structure. Furthermore, integration with geostationary satellite data enables lifecycle analysis, encompassing developing, mature, and dissipating stages. Research findings reveal several significant insights. Temporal and spatial variations: Convective intensity generally strengthens from early to mid-summer over southern China but weakens during the South China Sea monsoon onset. Regionally, convection is strongest over South China, followed by the Yangtze-Huaihe Valley, the Tibetan Plateau, and the East China Sea. The dry environment over the Tibetan Plateau leads to systems with high cloud bases, smaller horizontal scales, active mixed-phase microphysics, and lower lightning rates. Synoptic and organizational influences: System characteristics are strongly modulated by synoptic conditions. Over the eastern China plains, extremely wide convective systems occur mainly under trough, subtropical high, or typhoon influences, each associated with distinct seasonal peaks and severe weather propensities. Different organizational modes (e.g., trailing stratiform, bow echo) exhibit contrasting convective intensities, with bow-echo systems being the most vigorous. Notably, satellite observations have revealed frequent linear MCSs near the Wuyi Mountains that were not undetected by ground-based radars. Lifecycle evolution: Combined GPM and geostationary satellite data show systematic lifecycle transitions—developing stages exhibit the largest convective fraction, mature stages display the greatest precipitation area and particle concentration, and dissipating stages are characterized by smaller particle sizes. Severe weather linkages:Lightning:Thunderstorms contribute 40%—50% of annual rainfall and 70%—80% of heavy rainfall (>20 mm·h^-1) in key regions. Lightning frequency correlates more strongly with the mixed-phase layer ice volume (35 dBZ echo volume) than with echo-top height. Extreme precipitation:Approximately 20%—50% of extreme precipitation events, particularly in coastal monsoon areas, are associated with weak convection. These systems are small in scale and shallow in depth, with minimal lightning activity. Their extreme rainfall is primarily driven by enhanced warm-rain processes (efficient collision-coalescence), as evidenced by sharp reflectivity increases below the melting level. Persistent challenges arise from satellite limitations. The relatively low temporal resolution necessitates synergistic use with other observational data, while the narrow radar swaths introduce truncation effects that bias statistical analyses. In addition, standard parameterization methods struggle to capture complex system morphologies. Future research should focus on 1) enhanced multi-source data fusion through space-air-ground integrated observation networks; 2) improved classification algorithms based on machine learning; 3) mechanistic investigations combining satellite climatologies with numerical modeling to diagnose governing processes;and 4) full utilization of new satellite capabilities, such as the wider swath of FY-3G. In conclusion, precipitation measurement satellites have fundamentally advanced our understanding of precipitation systems and convective weather in China. Continued progress through multi-source integration and emerging technologies will further elucidate system complexities, ultimately improving precipitation prediction and disaster mitigation capabilities.

Abstract:Accurate knowledge of the phase of small cloud particles (d<50 μm) is crucial for understanding cloud microphysical processes and radiative effects,which remain major sources of uncertainty in weather and climate models.In situ measurements provide an effective means of obtaining cloud particle phase information;however,detecting the phase of small particles in airborne cloud physics observations has long been challenging.Utilizing the polarization properties of cloud particles to distinguish their phase has recently become recognized as an effective approach.This study focuses on the processing and application of polarization parameters measured by the Cloud Aerosol Spectrometer with Depolarization (CAS-DPOL),a commercial instrument developed by Droplet Measurement Technologies (DMT),USA.The CAS-DPOL was recently introduced into China's meteorological observation system and developed—together with other cloud microphysical probes—aboard the Inner Mongolia weather modification aircraft Y-12.Despite its operational use,no prior studies in China have reported the application of polarization information from CAS-DPOL,primarily due to the absence of dedicated data processing methods.
To address this limitation,this study integrates and refines a set of algorithms and threshold criteria suitable for processing CAS-DPOL polarization parameters to retrieve cloud particle phase states and related microphysical quantities.Additionally,the inter-arrival time (IAT) and transit time (TT) of detected particles are analyzed to identify shattered and coincidence particles,respectively.
Based on these methods and thresholds,the CAS-DPOL data collected during an airborne campaign over Northeast China,23 May 2021—under a cold vortex synoptic pattern—were analyzed.The results show that,the temperature in the detected clouds was between -1 ℃ and -8 ℃,supercooled cloud droplets accounted for approximately 80.18% of the total particle number concentration.The cloud phase was predominantly mixed-phased,although some regions consisted entirely of supercooled water clouds.For this weather system,a particle number concentration greater than 5 cm^-3 was found to be a reliable lower threshold indicating the presence of supercooled droplets,providing a practical reference for determining cloud-seeding suitability.Furthermore,analysis of in-flight natural icing intensity indicated only minimal icing,with no conditions meeting the CCAR-25 Appendix C criteria.The preliminary results demonstrate that supercooled liquid droplets can be effectively distinguished from ice crystals using the proposed processing methods in conjunction with CAS-DPOL measurements.Although further validation is needed,the findings highlight the strong potential of CAS-DPOL and the developed algorithms for applications in cloud microphysics research,artificial weather modification,and aircraft icing detection.

Abstract:This study investigates the impacts of five aircraft-based artificial rain enhancement operations conducted across Jiangsu Province during 3—9 November 2023 on local and regional air quality.Flight operation records were integrated with data from the Jiangsu provincial meteorological observation network and concurrent hourly environmental monitoring of particulate matter (PM_2.5 and PM₁₀) and gaseous pollutants (SO₂,NO₂,O₃,and CO) to evaluate both precipitation enhancement effectiveness and its influence on atmospheric pollutant concentrations.Periods affected by seeding were identified through the spatial and temporal overlap between aircraft flight tracks and the targeted cloud regions,along with the estimated downwind influence window.Monitoring sites located within these regions were designated as “impacted stations”,while a control area comprising control stations—chosen to represent similar meteorological and environmental conditions but unaffected by seeding—served as a baseline for comparison.
For each operation,pollutant concentrations and their change ratios during the seeding period were computed and contrastedbetween impacted and control stations.This comparative approach helps isolate the influence of seeding-induced precipitation from background variability due to synoptic changes or emission fluctuations.Results show that artificial precipitation enhancement significantly increased observed rainfall and was accompanied by marked reductions in particulate concentrations,with effects persistingseveral hours after rainfall onset.In the representative case on 3 November,PM_2.5,PM₁₀,and O₃ mass concentrations decreased for approximately 4—7 hours,indicating that wet scavenging processes remained active during and shortly after precipitation.Across the five operations,all monitored pollutants exhibited varying degrees of decline,with the most consistent and pronounced reductions observed for PM_2.5,PM₁₀,and O₃.Peak reductions during periods of intense precipitation reached approximately 90％,while more moderate events also yielded notable declines relative to pre-rainfall baselines.
Compared to gaseous pollutants,particulate matter displayed a stronger and more consistent response to precipitation.Both the absolute decreases and normalized change ratios of PM_2.5 and PM₁₀ were larger at impacted stations than at control stations,with differences amplifying as precipitation duration increased.This dependence suggests enhanced removal efficiency under prolonged rainfall,consistent with established wet-scavenging mechanisms.Specifically,in-cloud nucleation scavenging removes aerosols that activate into cloud droplets,while below-cloud collection processes—impaction,interception,and,diffusion—efficiently eliminate remaining aerosols as raindrops descend through the boundary layer.Gaseous pollutants exhibited weaker and more variable responses,reflecting differences in solubility,reactivity,and boundary-layer dynamics.
Uncertainties remain due to the limited sample size (five operations over one week) and potential confounding from concurrent synoptic variability,spatial heterogeneity in emissions,and boundary-layer evolutions.Nevertheless,converging evidence—including temporal alignment between rainfall and pollutant declines,stronger effects at impacted sites,and dependence on precipitation duration—indicates that aircraft-based cloud seeding can provide short-term but measurable air-quality benefits through enhanced wet removal of PM_2.5 and PM₁₀.These findings underscore the potential co-benefits of precipitation enhancement for pollution mitigation and highlight the need for expanded,multi-seasonal analyses supported by high-resolution microphysical and dispersion modeling.

Abstract:Previous studies have shown that, compared with direct model outputs, the accuracy of multi-model ensemble forecasts can be significantly improved. However, many existing approaches have limitations in representing forecast uncertainty and balancing skill with operational practicality. In East China, numerical models generally perform better than in other regions of the country, highlighting the need to develop techniques that can further enhance forecast performance.A multi-model ensemble technique,the recursive Bayesian model process (RBMP), has recently been developed at the NCEP-EMC to produce calibrated probabilistic forecasts. RBMP introduces two key modifications to the traditional Bayesian model averaging (BMA) method. First, the iterative parameter estimation process, typically performed using the expectation-maximization (EM) algorithm, is reformulated into a recursive procedure using a decaying-average method. This approach reduces the demand for data storage and improves adaptability to model upgrades,producing forecasts termed decaying-averaged BMA (DCBMA). Second, a decaying-average-based second-moment adjustment is applied to DCBMA-calibrated forecasts to correct over- or under-dispersed ensembles, resulting in the final RBMP forecasts.
To assess the operational utility of RBMP, the technique was applied to post-process ensemble predictions from three major global models—the ECMWF, NCEP, and CMC—using surface observations from 2016 to 2017.Bias correction was first performed for each model using the decaying-average method.The skill of the ensemble forecasts was then evaluated by comparing bias-corrected single-model forecasts, equal-weighted multi-model ensembles, and RBMP forecasts. Sensitivity tests on the decaying weight were also conducted, and an implementation scheme for 2 m temperature forecasts in East China was developed.In addition, RBMP was compared with traditional BMA under identical data conditions,and its applicability to failure cases was preliminarily examined using ECMWF and NCEP forecasts from September 2019 to May 2020 in Anhui Province.
The results demonstrate that RBMP provides more reliable probabilistic forecast distributions and improves forecast skill, particularly at short lead times. RBMP-based multi-model ensembles outperform both bias-corrected ECMWF ensembles and equal-weighted multi-model ensembles, reducing the RMSE of the ensemble mean by 3.0％—10.5％ and 2.0％—5.0％ in winter,respectively. RBMP also enhances resolution for both high-and low-temperature events, improves forecast accuracy in most failure cases, and provides valuable uncertainty information for difficult forecasts. Overall, RBMP retains the strengths of BMA while being more computationally efficient and requiring less storage. Moreover, ensemble spread is effectively calibrated through second-moment adjustment,making RBMP a practical method for improving short-range temperature forecasts.
A limitation of RBMP is that rapid changes in model performance, such as during seasonal transitions, may reduce its stability and negatively impact results.Future work should investigate whether RBMP can be extended to non-normally distributed variables such as wind speed and precipitation.

Abstract:The accuracy of sea surface height (SSH) merged maps is critical for monitoring oceanic small-scale variability, which underpins both oceanographic and meteorological applications. The integration of satellite radar altimeter data has substantially improved the quality of these maps. The deployment of China's autonomous Haiyang-2 (HY-2) satellite series has enhanced the country's independent ocean observation capacity while also contributing to the global ocean observing system. This study evaluates the impact of incorporating HY-2 data on the accuracy of SSH merged maps, a topic that has received limited attention to date. The combined contribution of indigenous satellites to SSH observations is of particular interest, as it offers potential improvements in the characterization of ocean dynamics and the accuracy of SSH predictions.
The primary objective of this study is to quantify the effect of HY-2 satellite data on SSH merged maps and to assess the resulting improvements in observational quality. Between 29 April 2022 and 3 February 2023, a two-dimensional variational merging method was applied to integrate SSH data from six international satellites and three HY-2 satellites. The resulting SSH merged maps, produced over the Northwest Pacific, had a spatial resolution of 0.12°×0.12° and daily temporal resolution. Accuracy was assessed through error statistics, correlation, and regression analyses, with cross-validation against geostrophic flow-corrected drifting buoy velocity data and tide gauge SSH observations. Results showed SSH errors of 2—5 cm, with flow velocity errors of 11 cm·s^-1 (zonal) and 13 cm·s^-1 (meridional), consistent with international merged products.
Inclusion of HY-2 satellites improved performance compared with buoy data, with vector direction errors reduced by nearly 0.1° and velocity root mean square error (RMSE) reduced by 0.31 cm·s^-1 (meridional) and 0.17 cm·s^-1 (zonal). Grid points with reduced RMSE accounted for 7% and 5% of the total buoy-covered area in the meridional and zonal directions, respectively. Against tide gauge observations, mean SSH error decreased by 0.1 cm and RMSE by 0.2 cm, while regression coefficients and correlation increased by 3%—5%. However, improvements plateaued once the number of merged satellites exceeded five, indicating a saturation point in observational density at approximately 50%. These results confirm that the inclusion of HY-2 satellites enhances the accuracy of SSH merged maps.
The HY-2 constellation plays a key role in complementing international data and supports diverse applications, including marine disaster prevention, transportation, resource development, environmental protection, scientific research, and national defense. Although traditional radar pulse technology has inherent limitations, merging altimetry data from three or more satellites effectively increases SSH map coverage and accuracy. Nevertheless, improvements diminish beyond a certain number of satellites, highlighting a point of diminishing returns. Future work should focus on optimizing merging algorithms and assessing the benefits of expanded satellite constellations for finer-resolution ocean monitoring. The findings provide valuable insights for strategic planning of satellite deployments and the utilization of merged SSH products in prediction models.