SEAPODYM-LMTL, the low and mid trophic level component of SEAPODYM, simulates mesozooplankton and micronekton biomass globally as an advection-diffusion-reaction system driven by physical and biogeochemical forcing. Its mesozooplankton parameterization remains incompletely calibrated, and its operational implementation couples the biological equations to spatial transport, so evaluating a parameter set requires running the full spatial model, which makes automated calibration costly. We present SeapoPym v0.1, an open-source Python re-implementation of the SEAPODYM-LMTL biological model that solves the dynamics locally, without transport. We apply it to the single epipelagic mesozooplankton group and estimate its five biological parameters with a covariance matrix adaptation evolution strategy (CMA-ES). Comparison with the operational product shows that omitting transport matters most in strongly advective regions and in cold high-latitude waters, where the long zooplankton life cycle keeps the biomass exposed to advection. At six contrasting stations the difference between the two models stays between 6 % and 12 % of the simulated biomass, five to seven times smaller than the model-observation gap where in-situ records allow that comparison. A Sobol analysis attributes the magnitude of the biomass to the energy-transfer and mortality parameters, and the timing of the seasonal peak to the recruitment parameters. Twin experiments then show that parameter identifiability depends on the environmental regime sampled. Wherever the search converged, energy transfer and mortality were recovered, whereas recruitment was recovered only in cold water. A single cold station constrained all five parameters as well as the six stations combined, so recovery follows the information content of the sampled regime rather than the number of stations. These results hold for noise-free synthetic observations generated by the transport-free model itself and driven by the exact forcing. The next step is to repeat them under realistic sampling and forcing error, then calibrate the model against in-situ records.
Data assimilation is used in a number of geophysical applications to optimally integrate information from observations and models. Providing an estimation of both state and uncertainty, ensemble algorithms are among the most successful data assimilation approaches. Since the estimation quality depends on the ensemble, the sampling method is a crucial step in ensemble data assimilation. This work introduces a sampling method featuring a higher polynomial order of approximation, and an ensemble filter, the Gauss-Hermite High-Order Sampling Hybrid filter (GHOSH), which exploits the higher order of the novel sampling method. In contrast, the order of the most frequently adopted ensemble algorithms in geosciences is usually equal to or lower than 2. In the directions where the uncertainty is larger, the GHOSH filter's sampling method achieves a higher order of approximation than in other ensemble-based filters, without increasing the asymptotic computational complexity that is comparable to that of second-order deterministic filters. To evaluate the benefits of the higher approximation order, a set of twin experiments of Lorenz96 simulations has been carried out using the GHOSH filter and a second-order ensemble Kalman filter (SEIK; singular evolutive interpolated Kalman filter). The twin-experiment results show that GHOSH outperforms SEIK in most of the assimilation settings, with up to a 56 % reduction of the root mean square error on assimilated and non-assimilated variables when best-tuned forgetting factors are adopted for each filter.
Faithful numerical representation of the coupled groundwater (GW)–unsaturated-zone (UZ)–surface-water (SW) dynamics at regional scales is constrained by the structural extremes currently available: fully three-dimensional Richards-equation solvers resolve variably saturated flow rigorously but at a per-evaluation cost that forecloses ensemble calibration and multi-decadal projection, whereas conceptual bucket schemes cannot track a moving groundwater head, represent capillary rise, or impose a soil-moisture-dependent evapotranspiration (ET) stress. This study introduces GWSWEX (Groundwater–Surface Water EXchange), a vertically resolved, process-based UZ model designed to sit between an external GW model and an external SW model within an integrated modelling chain. Its principal innovations are two interchangeable solvers behind a single unified API: an explicit operator-split bucket-sequence solver with CFL-adaptive sub-stepping and an implicit mixed-form Richards solver with Picard linearisation, sharing the same Mualem–van Genuchten constitutive relations and piecewise-linear drainable-volume function for a freely moving groundwater head; the per-step storage budget closes to floating-point tolerance in the explicit solver and to a small, well-characterised residual in the implicit solver, enabling a coupled GW–UZ–SW iteration to converge without under-relaxation. Verification against HYDRUS-1D across six soil profiles under two contrasting forcing scenarios shows sub-centimetre to near-centimetre groundwater-head accuracy under smooth daily forcing, with eleven of twelve basic-scenario cases attaining an RMSE at or below the 2 cm discretisation step and a wet-phase mean RMSE of approximately 4 cm (implicit) and 8 cm (explicit) under intensive hourly forcing; an iterated one-at-a-time sensitivity analysis characterises the empirical parameters and the structural limits of the layered-bucket abstraction, while a parallel-ensemble benchmark shows both solvers completing a thousand-column ensemble six to twelve times faster than parallelised HYDRUS-1D, at sub-linear run-time scaling exponents of approximately 0.92–0.93 that render ensemble-grade calibration and multi-decadal projection tractable on mainstream consumer-grade hardware.
To our knowledge, this study presents the first implementation of an observationally-constrained Generalized Double-Moment scaling Normalization (GDMN)-based rain Drop Size Distribution (DSD) representation within Weather Research and Forecasting (WRF) Double-Moment 6-class (WDM6) scheme and evaluates its impacts in a convection-permitting real-case simulation. The modified scheme, referred to as WDM6-GDMN, is evaluated through simulations of an isolated summer convection case over the Korean Peninsula, using the universal double-moment normalized DSD function, h(x), derived from rain DSDs observed in the Boseong region during the summers of 2018 and 2019. WDM6-GDMN provides a more realistic spatial distribution of surface precipitation by better simulating convection-cell movement. Although none of the cloud microphysics parameterizations, including the bin-type scheme, reproduce the observed convection that developed in the southeast of the analysis domain, only WDM6-GDMN successfully captures this feature. Microphysical analysis demonstrates that, in WDM6-GDMN, enhanced cloud production due to stronger upward motion leads to the formation of more raindrops and, consequently, greater surface precipitation over southeastern region. Furthermore, the contoured frequency by altitude diagrams of radar reflectivity for the WDM6-GDMN reveals slower particle growth and weaker reflectivity in the lower atmosphere compared with the original scheme, in better agreement with observations.
This paper outlines the scope, development and publication of data requirements for a set of reference climate simulations, and describes the methodology used to gather and synthesise them into a cohesive “Data Request” usable by data producers. The simulations supported by this Data Request comprise the initial phase of the next upgrade of research activities under the World Climate Research Programme (WCRP). Dunne et al. (2025) set out the scientific scope and objectives of the Assessment Fast Track and its role in initiating the current phase of the Coupled Model Intercomparison Project, CMIP7 – a key part of the next WCRP activity upgrade. Building on the successes of past CMIP phases which have supported an ever-expanding scope of work, the Data Request Task Team reached out to new communities to enhance engagement in CMIP. A transparent and community-led approach was adopted, where domain experts from the CMIP community were recruited into five teams by domain called “Thematic Author Teams” to co-create data requirements through wide consultation. This paper describes the process of gathering data requirements in the initial phase of CMIP7, along with the structure of the CMIP7 Data Request itself. Version 1.2.2.5 of the CMIP7 Data Request (DR7) covers the data production requirements for (i) control simulations of the past climate (both distant and recent), (ii) key sensitivity experiments focusing on critical aspects of climate and model behaviour, and (iii) a range of future climate scenarios. It consists of a relational database that maps climate variables to reference experiments according to scientific objectives, with associated metadata to enable data production and tools to allow interoperability and content exploration. The usage of climate models and CMIP data is broadening from its origins in scientific study of the physical environment to support the analysis of climate impacts, and planning for an ever-increasing portfolio of adaptation and mitigation measures. To support this growing scope, DR7 introduces a new organising component, Opportunities, to support transparent mapping between variables and experiments. The 46 Opportunities in DR7 represent the key community-driven use cases across CMIP data users – each describing why its combination of variables and experiments is important and how they contribute to impact, providing both scientific justification and technical requirements. It is challenging to represent the needs of the rapidly expanding CMIP community and user base while respecting the capacity limitations of CMIP data production. DR7 addresses this through a process of wide stakeholder engagement centred around an open consultation and community co-creation, while prioritising stakeholder representation and diversity. Innovations in interactive web tools and enhanced WCRP support through the CMIP IPO and Task Teams were also critical to the process of developing data requirements in collaboration with the community.
Due to their wide variety of properties, the representation of ice particles in cold and mixed-phase clouds are challenging to represent for microphysical schemes. To improve their representation, this study evaluates the implementation of predicted rime mass distribution in the bin microphysics scheme DESCAM. Based on the “fill-in” concept, the model allows a smooth transition in ice particle properties between unrimed and graupel particles. Consequently, the terminal velocity and collision kernels of ice particles were updated as a function of rime fraction. First, this study investigates the impact of the new implementation on cloud microphysics for an idealized squall-line system. The results show that large ice particles acquire substantial rime mass, enhancing their sedimentation and leading to earlier and more intense precipitation (+15 %), while weakening the cold pool, reducing buoyancy, and slowing squall-line propagation. The second part of this study is dedicated to evaluating the new version of DESCAM against field observations from the ICE-POP 2018 campaign, focusing on a heavy snowfall event that occurred from 7–9 March. For this case, we found that the simulated rime mass fraction at ground evolves similarly to the rime index measured by the MASC instrument. The new DESCAM version also produces significant changes in the amount and spatial distribution of precipitation, with strong local variations exceeding 10 mm and a 7.2 % increase in total accumulation, which leads to a better agreement with field observations. Overall, accounting for predicted rime mass improves consistency between the model and ground-based observations from ICE-POP 2018.
The increasing availability of kilometer-scale climate simulations presents major challenges for data access, processing, and analysis due to the unprecedented volume and heterogeneity of the outputs. Different data formats, structures, and metadata conventions, require dedicated solutions to ensure interoperability and usability. We introduce AQUA (Application for QUality Assessment), a Python-based framework developed within the Climate Change Adaptation Digital Twin of the Destination Earth (DestinE) initiative, designed to support the automated evaluation of high-resolution global climate simulations. Although several diagnostic suites for the analysis of global climate model data are already available, AQUA provides a flexible and modular infrastructure for accessing and processing climate model output across various formats. By building on widely adopted Python libraries, it enables scalable, out-of-core computations. Its design supports integration into automated workflows and user-defined pipelines, facilitating both operational and research-oriented applications. This paper focuses on the architecture and core functionalities of the AQUA core, which handles data ingestion, standardization, and pre-processing. AQUA is open source and actively maintained, and aims to serve as a community tool for robust, reproducible, and efficient climate data analysis across projects and institutions.
Global storm-resolving simulations with kilometer resolution are increasingly replacing traditional climate modeling approaches and show particular potential for resolving the dynamics and effects of deep moist convection. These modern modeling methods are moving toward sub-km scales, leading to extremely high energy and resource requirements. This makes iterations, parameter optimizations, and sensitivity studies no longer easily feasible. For the class of propagating, large-scale weather phenomena such as hurricanes, high-resolution limited-area approaches in combination with Lagrangian methods are therefore of interest, in which refined grids are shifted along the path of the phenomenon under consideration. To create this capacity for the ICON atmospheric model, this study develops a flexible workflow toolkit to enable efficient simulations of hurricanes on a sub-km scale. Our approach leverages the flexibility that ICON offers through the ability to create custom grids. The concept divides hurricane tracks into overlapping temporal windows of 12–24 h and generates customized grid segments that follow the hurricane's movement. The technical implementation automates key components of the workflow, including hurricane tracking, flexible grid generation, and preparation and merging of initial conditions across successive segments. The application of the workflow is demonstrated using Hurricane Paulette (2020) as an example, for which high-resolution simulations with grid spacings down to 300 m were performed using different segment configurations. The results show that the hurricane track remains consistent with the base run and depends mainly on the across-track width of the chosen configuration, while intensity metrics such as minimum pressure and maximum wind speed show significant sensitivities to resolution in our multi-nested setups. The efficiency gains are significant compared to traditional approaches with fixed limited-area domains: replacing a fixed domain with a track-following tube reduces resource requirements by factors of 2–8, and segmenting the tube into short, frequently re-initialized intervals adds a further factor of 2–6. Analysis of re-initialization effects shows systematic but manageable impacts during segment transitions. Nevertheless, the efficiency gains achieved by our method are so substantial that they justify the acceptance of the re-initialization effects. Our segment-based approach in the hurricane-centric reference system now allows for more flexible regional hurricane simulations with the ICON model and more efficient investigation of new research questions regarding the sensitivity of hurricane cloud systems at very high resolutions.
Abstract. A 21-member ensemble of regional climate simulations has been produced for Southeast Asia (SEA) by dynamically downscaling Coupled Model Intercomparison Project Phase 6 (CMIP6) Global Climate Models (GCMs) under the World Climate Research Programme's Coordinated Regional Climate Downscaling Experiment (CORDEX). The ensemble was generated by several modelling institutes using three regional climate models (RCMs) with eight distinct model configurations, resulting in a total of 62 simulations spanning the historical period and multiple future emissions scenarios. Model performance for mean, daily maximum/minimum temperature, and precipitation was evaluated against multiple observations at annual, seasonal, and daily time scales over SEA and its two subregions: Mainland and Maritime Continent (MC). Despite large observational uncertainties in precipitation intensity, the CMIP6 CORDEX-SEA ensemble captures the spatial and seasonal rainfall distribution reasonably well but tends to substantially overestimate observed rainfall. Wet biases, evident in about two-thirds of the models, are regionally and seasonally heterogeneous and larger over monsoon-dominated regions and seasons (e.g., MC during November–April and the Mainland during May–October). All RCMs showed widespread, statistically significant cold biases in daily mean temperature, which were largest during boreal winter, over the Mainland, and in simulations that have significant wet biases. These cold biases primarily arise from the models' underestimation of daily maximum temperature. The MC remains a challenging region since models struggle to accurately capture the spatial variability of rainfall and the internal variability of temperature. A standardised benchmarking framework was applied to precipitation and temperature, which ultimately identified 15 historical simulations that met our a priori model performance expectations. Analysing the range of future projections and model independence shows that simulations from the same RCM family exhibit similar bias structures, highlighting the importance of RCM setup and the selection of statistically independent models. From this process, eight simulations spanning three RCM configurations were selected for further kilometre-scale dynamical downscaling over megacities of SEA.
The concentrations of the major cations (esp., Ca2+, Mg2+) in Earth's oceans have undergone large-scale fluctuations in the geological past. This is important because the key geochemical properties of the marine environment that underpin the global carbon cycle – the aqueous carbonate system equilibria and solubility of solid calcium carbonate (CaCO3) – are heavily influenced by ion-pairing, which in turn depends on the activity of the major cations and anions. An appropriate interpretation of marine proxies as well as the reconstruction of past states of ocean geochemistry and carbon cycle across geologic events requires that these effects are considered. However, most current global carbon cycle models use empirical carbonate system dissociation constants fitted to laboratory experiments with present-day seawater major cation and anion concentrations and in the few simulations of global carbon cycling that have accounted for paleo-seawater composition, only relatively simplified empirical adjustments of the equilibrium constants (from Ben-Yaakov and Goldhaber, 1973; Tyrrell and Zeebe, 2004) have been implemented (e.g., Panchuk et al., 2008). Here we develop and evaluate a new scheme in the cGENIE Earth system model for correcting carbonate system equilibrium constants and accounting for variations in the dissolved calcium and magnesium concentrations in the ocean. We base our new parameterization on the MyAMI specific ion interaction model of Hain et al. (2015) and implement this in cGENIE by means of linear interpolation within a 4-dimensional parameter look-up table of pre-calculated carbonate system equilibrium constant values. For modern seawater composition, our implementation of MyAMI-based equilibrium constants yields no meaningful deviation from model results using empirically-based equilibrium constants, validating our look-up/interpolation approach. However, for simulations conducted under non-modern Mg/Ca, we find substantial differences in carbon chemistry and CaCO3 saturation when using our new MyAMI-based equilibrium constants as compared to the existing (default) correction scheme in cGENIE. Specifically, our new MyAMI-based correction scheme exhibits a much lower sensitivity to an instantaneous change in Mg/Ca from modern to Eocene and an approximate doubling of Ca2+ concentration, in both surface ocean pH (0.01) and calcite saturation state (9.41), which were overestimated by 0.05 and 4.04, respectively, with the previous correction scheme. Any bias in carbonate chemistry and CaCO3 saturation will also affect the preservation and burial of CaCO3 in deep-sea sediments and hence potentially impact model-data comparisons. We illustrate this by contrasting the ocean carbon inventory arising under Eocene Mg/Ca with the same total weathering (and hence CaCO3 burial) flux for the different possible equilibrium constant corrections. We find that the new MyAMI-based and previous default corrections give rise to a dissolved inorganic ocean carbon inventory 20 µmol kg−1 (348 PgC) higher and 59 µmol kg−1 (950 PgC) lower, respectively, relative to the same experiment conducted using empirical equilibrium constants without any Mg/Ca correction. Applying no correction at all for a different-from-modern Mg/Ca ratio in the ocean would appear to be better than applying an overly-approximated correction but explicitly accounting for past dissolved calcium concentrations remains of fundamental importance. We provide this new carbonate system equilibria correction as an option in cGENIE.muffin version 0.9.76, and as standard in a forthcoming new cGENIE code release – cGENIE.cookie v.0.91.
In this study, a wave module is online-coupled into the Parallel Ocean Program version 2 (POP2) within the CESM1.2.2 framework (hereafter POP2–waves). Unlike empirical data analyses of observation-based products and offline-forced ocean models that treat wave-induced gas transfer velocity and the air–sea difference in partial pressure of CO2 (ΔpCO2) as decoupled variables, the online-coupled POP2–waves model captures their interactive, ΔpCO2-mediated negative feedbacks. This setup enables wave properties to dynamically modulate gas transfer velocity and physical mixing through sequential processes. POP2–waves is evaluated alongside a control simulation (B–CTL) against the National Oceanic and Atmospheric Administration (NOAA) CarbonTracker (CT2022) inversion product. The spatial air–sea CO2 flux in POP2–waves simulation shows generally closer structural agreement with NOAA CT2022 than the control B–CTL, thereby improving performance in most selected regions. Specifically, bubble-mediated transfer accounts for up to 41.3 % of the total flux, consistent with the ∼40 % contribution reported in recent research. Although the inclusion of waves (POP2–waves) enhances regional oceanic CO2 uptake by 11.8 % and outgassing by 41.6 %, these two processes largely offset each other. Consequently, this dual enhancement results in only a slight 1.8 % increase in the global net ocean CO2 sink compared to the B–CTL simulation. Globally, air–sea ΔpCO2 and pH exhibit the strongest positive and negative regression coefficients with the air–sea CO2 flux, respectively. Regionally, the gas transfer velocity shows a positive (negative) regression coefficient within oceanic CO2 outgassing (uptake) regions, whereas SST displays the opposite trend.
Abstract. Quantifying greenhouse gas (GHG) emissions over complex terrain remains a significant challenge for conventional inversion systems due to the high sensitivity of tracer transport to surface heterogeneity. We develop a high-resolution dual-species GHG top-down inversion framework by integrating the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem v4.4) and the Data Assimilation Research Testbed (DART v9.8.0) in a cycling ensemble Kalman filter system. Built on community tools, the framework is designed to be portable and configurable, enabling applications to other regions, resolutions, and observing-system configurations (e.g., expanded surface networks and additional data streams). This framework jointly assimilates near-surface CO2 and CH4 concentrations to produce dynamically consistent updates of emissions. By employing a unified Eulerian framework that simultaneously updates meteorology and 3-D tracer fields, the system is designed to maintain consistency between transport and concentrations and to reduce the risk that transport-related concentration errors are aliased into flux adjustments over the complex landscapes of the Korean Peninsula. To improve the simulation of turbulent GHG dispersion in the atmospheric boundary layer over complex terrain, we incorporate surface heterogeneity parameterizations (roughness sublayer and canopy height) into the model physics in the inversion system. The system assimilates high-precision continuous in situ observations from three World Meteorological Organization/Global Atmosphere Watch (WMO/GAW) stations to constrain CO2 and CH4 emissions. Prior flux estimates include anthropogenic emissions from the Emissions Database for Global Atmospheric Research (EDGAR v8.0), biogenic exchanges (the region-optimized Vegetation Photosynthesis and Respiration Model), biomass burning (Fire Inventory from the National Center for Atmospheric Research v2.5), and oceanic CO2 exchanges (SeaFlux). In a 2020 case study, the top-down estimates improve the agreement with ground observations, reducing root-mean-square errors by 30 %–60 % and lowering posterior mean bias to 1–2 ppm for CO2 and 20–30 ppb for CH4 at the high-precision surface observation sites. Independent aircraft profiles provide external evaluation and indicate residual CH4 discrepancies consistent with prior emissions and boundary condition uncertainties. Controlled observing-system simulation experiments show that, under prescribed perturbations, the system produces bounded and interpretable emission responses to transport-model, boundary-condition, and observation-error perturbations. They also indicate that recovery is strongest within station footprints and remains coverage-limited under the current three-station network, while a dense-network known-truth experiment highlights the value of expanded observational coverage for improving domain-wide emission constraints. Posterior adjustments suggest reduced CO2 emissions over the Seoul Metropolitan Area and parts of the western coastal region and increased CH4 emissions over inland agricultural source regions relative to the priors, highlighting priorities for follow-on evaluation of inventory components. Our posterior CO2 total (620 ± 45 Mt yr−1) is consistent with the Republic of Korea Biennial Transparency Report (ROK-BTR) estimate (624 Mt yr−1) at the national scale, while the posterior CH4 total (54.7 ± 5.2 Mt CO2eq yr−1) exceeds the ROK-BTR estimate (35.5 Mt CO2eq yr−1) by 19.2 Mt CO2eq yr−1, consistent with the larger structural uncertainty in CH4 source characterization and spatial allocation noted in previous studies.
Quantifying greenhouse gas (GHG) emissions over complex terrain remains a significant challenge for conventional inversion systems due to the high sensitivity of tracer transport to surface heterogeneity. We develop a high-resolution dual-species GHG top-down inversion framework by integrating the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem v4.4) and the Data Assimilation Research Testbed (DART v9.8.0) in a cycling ensemble Kalman filter system. Built on community tools, the framework is designed to be portable and configurable, enabling applications to other regions, resolutions, and observing-system configurations (e.g., expanded surface networks and additional data streams). This framework jointly assimilates near-surface CO2 and CH4 concentrations to produce dynamically consistent updates of emissions. By employing a unified Eulerian framework that simultaneously updates meteorology and 3-D tracer fields, the system is designed to maintain consistency between transport and concentrations and to reduce the risk that transport-related concentration errors are aliased into flux adjustments over the complex landscapes of the Korean Peninsula. To improve the simulation of turbulent GHG dispersion in the atmospheric boundary layer over complex terrain, we incorporate surface heterogeneity parameterizations (roughness sublayer and canopy height) into the model physics in the inversion system. The system assimilates high-precision continuous in situ observations from three World Meteorological Organization/Global Atmosphere Watch (WMO/GAW) stations to constrain CO2 and CH4 emissions. Prior flux estimates include anthropogenic emissions from the Emissions Database for Global Atmospheric Research (EDGAR v8.0), biogenic exchanges (the region-optimized Vegetation Photosynthesis and Respiration Model), biomass burning (Fire Inventory from the National Center for Atmospheric Research v2.5), and oceanic CO2 exchanges (SeaFlux). In a 2020 case study, the top-down estimates improve the agreement with ground observations, reducing root-mean-square errors by 30 %–60 % and lowering posterior mean bias to 1–2 ppm for CO2 and 20–30 ppb for CH4 at the high-precision surface observation sites. Independent aircraft profiles provide external evaluation and indicate residual CH4 discrepancies consistent with prior emissions and boundary condition uncertainties. Controlled observing-system simulation experiments show that, under prescribed perturbations, the system produces bounded and interpretable emission responses to transport-model, boundary-condition, and observation-error perturbations. They also indicate that recovery is strongest within station footprints and remains coverage-limited under the current three-station network, while a dense-network known-truth experiment highlights the value of expanded observational coverage for improving domain-wide emission constraints. Posterior adjustments suggest reduced CO2 emissions over the Seoul Metropolitan Area and parts of the western coastal region and increased CH4 emissions over inland agricultural source regions relative to the priors, highlighting priorities for follow-on evaluation of inventory components. Our posterior CO2 total (620 ± 45 Mt yr−1) is consistent with the Republic of Korea Biennial Transparency Report (ROK-BTR) estimate (624 Mt yr−1) at the national scale, while the posterior CH4 total (54.7 ± 5.2 Mt CO2eq yr−1) exceeds the ROK-BTR estimate (35.5 Mt CO2eq yr−1) by 19.2 Mt CO2eq yr−1, consistent with the larger structural uncertainty in CH4 source characterization and spatial allocation noted in previous studies.
A hybrid framework for simulating SPontaneous synthetic tropical cyclones (TCs) with realistic INtensity, hereafter SPIN, is developed for TC risk assessment. Unlike earlier statistical synthetic-TC models, SPIN retains intra-month variability and provides the ambient environment for each storm. The SPIN model leverages a Neural General Circulation Model (NeuralGCM) to simulate spontaneously generated TC tracks, and then couples a dynamical TC intensity model to estimate their intensity evolution based on the large-scale environment. It reproduces key features of observed TC climatology, including interannual variability, seasonal cycle, genesis and track patterns, lifetime-maximum intensity, regional landfall-intensity return periods and inter-genesis intervals. It also improves the representation of interannual TC variability in basins where ENSO-related influences are spatially mixed. Beyond individual TC events, the model demonstrates improved skill in representing multiple tropical cyclone events (MTCEs), capturing their year-to-year variability and cluster size. SPIN advances TC risk assessment by enabling compound TC hazards, such as MTCEs and beyond, to be assessed as dynamically organized events rather than statistical co-occurrences alone.
This study provides a comprehensive description of the China Meteorological Administration Climate Prediction System version 4 (CMA-CPSv4), which is developed based on the fully coupled global climate-aerosol Beijing Climate Center Earth System Model (BCC-ESM1). It is updated from its previous version, CMA-CPSv3, which was based on the high-resolution Beijing Climate Center Climate System Model version 2 (BCC-CSM2-HR). In contrast to CMA-CPSv3, CMA-CPSv4 is capable of simulating the dynamic evolution of aerosols and their feedback on the climate system. This study aims to evaluate the reproducibility of atmospheric aerosols in CMA-CPSv4 under the forcing of observed atmospheric circulation. The 20-year simulations for the period 2001–2020 are conducted. The results show that CMA-CPSv4 reasonably captures the global spatial distribution and temporal variations in mass concentrations for five categories of dust, sea salt, sulfates, organic carbon, and black carbon, as well as aerosol optical depth (AOD). In East Asia, simulated fine-mode particulate matter PM2.5 concentrations are generally consistent with the CMIP6 multi-model ensemble mean (MME), although dust concentrations over the Taklamakan–Mongolia–North China regions are slightly underestimated, and sulfate concentrations are overestimated over the oceans. In addition, several severe dust pollution events in northern China are successfully reproduced, demonstrating the capability of CMA-CPSv4 to simulate aerosol concentrations and extreme pollution events. The reasonable simulation of aerosol distribution is fundamental for studying aerosol-climate interactions and the impact of aerosols on numerical weather and climate prediction in our future work.
This paper details a computationally efficient and versatile Python package (BinMod1D v1.0.10) that explicitly evolves spectral bin distributions and corresponding polarimetric radar variables for rain or snow according to atmospheric collisional coalescence and breakup processes. BinMod1D can be executed as a box model, a 1D steady-state model in height, or a full (time and height) 1D column model utilizing multiple particle categories, each of which can have their own densities, aspect ratios, and fall speeds. Forward simulations of polarimetric radar variables are implemented using standard Rayleigh analytic scattering equations. Two-moment (mass and number) or one-moment (mass only) particle interaction calculations follow a source-based spectral bin method. Bin interaction computations are parallelized using just-in-time (JIT) compilation for high performance. BinMod1D box model solutions are validated using analytic solutions of collision-coalescence using a variety of kernels as well as for breakup and the steady-state balance of coalescence with breakup. BinMod1D capabilities are demonstrated through steady-state simulations of rainfall and snow signatures, as well as vertical profiles of diverse meteorological scenarios. Convergence and timing tests are provided for the meteorological scenario of a cloud to rain transition using a realistic collision kernel and fragment distribution. BinMod1D is intended to enable cloud microphysics and weather radar researchers to efficiently simulate vertical profiles of complex weather events. Such a tool can be used to provide reference solutions for training machine learning models and validating various retrieval methodologies.
Arctic mixed-phase clouds (MPCs) remain challenging to represent in atmospheric models, with bulk microphysics schemes typically biased toward either excessive glaciation or inadequate ice formation. This study evaluates the behavior of the Weather Research and Forecasting (WRF) Double-Moment 6-class (WDM6) scheme and its modified version (WDM6_ICE), which incorporates spherical ice shape, constrained nucleation, and prognostic cloud ice number concentration, under Arctic conditions using the Mixed-Phase Arctic Cloud Experiment (M-PACE) case (9–10 October 2004). WDM6 severely underestimates liquid water through efficient vapor deposition onto bullet-shaped ice. WDM6_ICE exhibits the opposite behavior, maintaining persistent liquid as cloud ice deposition is reduced by about 98 %. The suppressed cloud ice is partially compensated by enhanced snow deposition, resulting in a reduction in total ice water content of about 70 % rather than complete ice elimination. Sensitivity experiments show that ice shape modification is the dominant factor, while nucleation modification alone redistributes ice among hydrometeor categories without reducing total ice content. Comparison with mid-latitude evaluations of the same scheme indicates a regime-dependent response, in which modifications that produce a moderate ice adjustment in mid-latitude cases lead to a more pronounced restructuring of the ice budget under Arctic conditions. Surface energy analysis indicates that balanced phase partitioning is more relevant than liquid water maximization for radiative bias reduction. All configurations underestimate total water path, suggesting that accurate Arctic MPC representation requires coordinated improvements in ice particle properties, ice-nucleating particle recycling, and boundary-layer coupling. These results are based on a single Arctic case and demonstrate a mechanism that remains to be tested across a broader range of conditions.
The three-dimensional unstructured-mesh finite-element atmospheric dynamical framework is gaining significance owing to its flexibility in representing complex topography and capability for multi-scale simulations in high resolutions. However, this framework has substantial bottlenecks. Unlike structured-grid models, the unstructured finite element method (FEM) must frequently access irregular mesh connectivity among nodes, edges, and elements, causing indirect memory addressing, inadequate data locality, and substantial memory bandwidth bottlenecks on conventional CPU architectures. Consequently, element-wise computations and global assembly are among the primary contributors alongside the sparse linear solver to the runtime in high-resolution simulations. This study develops a GPU-parallel implementation of the Fluidity-Atmosphere dynamical core to address these challenges. The GPU-oriented data structures and optimized kernels are designed to efficiently leverage the computing power of GPUs. These kernels enable parallelized element integration and are efficient solvers for specific size matrices; a parallel assembly strategy enhances memory throughput during global sparse matrix construction. On the NVIDIA A100 GPU, the optimized kernels achieve speeds over 100× compared to the single CPU core baseline for element-wise computations and up to 389.02 times for global matrix assembly, resulting in an overall acceleration of 8.57 times with four messages passing interface (MPI) processes. The proposed framework demonstrates that tailored GPU parallelization is effective in overcoming the computational bottleneck of unstructured FEM-based atmospheric models, facilitating high-resolution simulations on heterogeneous architectures.
GLIDE-SOL is a fully scripted and globally re-deployable Python workflow that operationalizes SOLWEIG for rapid and repeatable thermal-comfort mapping across diverse urban environments. GLIDE-SOL is built on the SOLWEIG radiative balance libraries, but rewrites the surrounding system – including automated input generation, the execution engine, and post-processing – so that the model can be driven by globally available datasets and executed efficiently on GPUs. All inputs (terrain, building morphology, canopy height, land cover, and meteorology) are automatically derived from global products, eliminating local preprocessing while enabling consistent applications from neighborhood-scale analyses to city-wide and multi-city experiments. In addition, GLIDE-SOL introduces lightweight physical diagnostics to improve realism when driven by coarse meteorological forcing, targeting key urban controls on wind and near-surface temperature. The workflow incorporates two physical augmentations: (i) roughness- and obstacle-based directional wind attenuation to approximate near-surface ventilation; and (ii) diagnostic temperature adjustments that combine a simple urban heat island (UHI) cycle with an elevation-based correction using high-resolution DEM information, to better capture nocturnal warming and local lapse-rate effects. To scale to large metropolitan areas, GLIDE-SOL uses explicit domain tiling with cross-tile synchronization to preserve radiative consistency across tile boundaries, enabling meter-scale simulations over tens to hundreds of square kilometers without sacrificing reproducibility. Daily outputs (24 radiative and meteorological fields) are stored as compressed GeoTIFFs to reduce disk usage and accelerate downstream processing. GLIDE-SOL is implemented through three reproducible components: an automated global-input generator; a SOLWEIG execution engine with coordinated tiling; and a post-processing module for systematic sampling, time-series extraction, and visualization. An operational demonstration in Dortmund, using hourly measurements from 25 urban and peri-urban stations and simulations run at 2 m grid spacing between August 2024 and December 2025, shows that incorporating wind attenuation and the diagnostic temperature corrections substantially improves UTCI performance (RMSE reduced from 8.09 to 2.84 °C), alongside improvements in Tmrt, air temperature, and wind speed simulations. By integrating harmonized global inputs with physics-based diagnostics, GPU acceleration, and scalable tiling, GLIDE-SOL supports applications such as operational UTCI nowcasting, retrospective and climatological analyses of heat stress, sensitivity tests of urban morphology and greening strategies, and coordinated multi-city experiments requiring consistent modeling protocols.
Regional sea ice models often do not cover the full extent of polar ice and instead include open ocean boundaries that are not ice-free year-round. This necessitates the specification of lateral boundary conditions for sea ice, an inherently challenging task for most sea ice models. Although this issue is less critical for pan-polar domains, the interior ice state still needs to be constrained for many applications. In this study, we present the design and evaluation of a Newtonian relaxation algorithm for sea ice, implemented in the NOAA Geophysical Fluid Dynamics Laboratory (GFDL) Sea Ice Simulator (SIS2). The algorithm can be applied both at the lateral boundaries to impose open boundary conditions and within the interior domain to constrain sea ice thickness and concentration toward prescribed target fields. The method is flexible and can be applied anywhere in the domain, making it especially well-suited for regional applications of sea ice models with variable ice cover along their boundaries. The method is evaluated within a regional forecasting system based on the NOAA GFDL ocean model (MOM6) coupled with sea ice model (SIS2) for two regional configurations: the Northeast Pacific and the Arctic Ocean. Sensitivity experiments spanning a range of relaxation time scales and nudging strengths demonstrate that the method substantially improves the representation of sea ice and associated ocean surface fields, offering a practical solution for both boundary and interior constraints in regional sea ice modeling.
