More of Earth’s surface is covered by stratocumulus clouds (Sc) than by any other cloud type, making Sc particularly important for Earth’s energy balance, primarily through the reflection of incoming solar radiation. However, representing Sc and their radiative impacts remains one of the greatest challenges for global climate modeling as models cannot resolve the length scale of the processes involved in its formation and evolution. For this reason, Sc represent a major source of uncertainty in climate projections (Wood 2012).
The challenge becomes more complicated due to the organizational complexity exhibited by Sc across a broad range of spatial scales. In particular, marine Sc fields display characteristic mesoscale patterns that can exhibit both organized and disorganized structures. Among these morphological regimes, cellular convection has received particular attention as cloud decks can self-organize into semi-regular tessellations composed of closed and open convective cellular fields.
Here we analyze satellite imagery of Sc organizing into low-reflectivity regions of open cells embedded within closed cellular cloud fields, known as "pockets of open cells" (POCs) (Stevens et al. 2005). We first track POCs from the time they emerge to when they dissipate. Second, a cell-scale analysis is performed for convective fields both inside and outside the POC boundaries to characterize the interface between POC and non-POC regions.
We propose a segmentation, tracking, and morphological analysis of cell geometry and dynamics in both closed and open cellular fields, with particular emphasis on the interactions between these cell types during POC development. A statistical analysis of multiple POCs is conducted to characterize the temporal and spatial contributions of cellular structural and kinetic changes to POC evolution, incorporating the local dynamics between individual cells (Farrell et al. 2017).
Using this framework, key differences between open and closed cellular regimes are identified based on velocity dynamics and morphological evolution.
Whereas closed cells exhibit relatively slow dynamics, open cells continuously rearrange, changing both size and shape, and display significant cellular mobility, revealing motion flows within the POC region.
Finally, shallow cold pools within POCs are identified based on their lifetime, area expansion, and interactions between neighboring cells. These cold pools, which result from stratiform precipitation from open cells, play a dominant role in the dynamics of open cell fields.
The primary result of this analysis reveals a previously unreported regime of collective cellular dynamics, in which the emergence and evolution of convective organization is strongly influenced by cold pools, exhibiting structural and dynamical behaviors not observed in other known cellular systems.
How to cite: Monroy Merida, D. L. and Haerter, J.: Morphological cellular analysis of Pockets of Open Cells on Marine Stratocumulus fields , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14944, https://doi.org/10.5194/egusphere-egu26-14944, 2026.
Rain processes are often underrepresented in our understanding of the trade-wind layer, despite trade cumuli precipitating ~30% of the time. The vertical structure of the main building blocks of precipitating convection, namely in-cloud updrafts and downdrafts, is poorly characterized due to a lack of suitable observations. Lidars cannot penetrate deeply into clouds, and hydrometeor fall speeds dominate the mean Doppler velocity from cloud-profiling radars. Here, we retrieve the vertical air motion inside precipitating clouds by making use of the Doppler velocity spectrum from the ground-based Ka-band radars at the Barbados Cloud Observatory (BCO), using methods previously developed for stratocumuli. The resulting dataset spans more than five years (2019-2025) at high (2s) resolution and is validated against available lidar measurements. We resolve circulations at the cloud scale. Shallow precipitating cumuli feature a narrow updraft at the cloud front that develops up to cloud top. The wider precipitation downdraft is triggered slightly below cloud top, where the rain content is large enough, and extends down to the surface. We show that deeper clouds are associated with stronger updrafts and downdrafts. Faster downdrafts are also associated with higher cloud reflectivity, suggesting that microphysical processes play a large role in determining their strength. The long record also lets us ascertain seasonal variability. Shallow precipitating cumuli feature faster updrafts in the summer trades, leading to larger and faster downdrafts than in winter. We show that the total mass flux of shallow precipitating cumuli is highly variable. Both the updraft and the downdraft mass fluxes mainly depend on the cloud fraction, but their balance hinges on the downdraft intensity. These observations can improve our understanding of tropical convection and shed light on the assumptions behind convective parametrizations and constrain cloud-resolving simulations.
How to cite: Poydenot, F., Robbins-Blanch, N., Zhu, Z., and Vogel, R.: Observations of vertical motion inside precipitating trade cumuli, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5727, https://doi.org/10.5194/egusphere-egu26-5727, 2026.
The maximum vertical velocity within deep convective updrafts (wmax) is a key control on precipitation intensity and lightning, yet the physical processes that set its magnitude remain unclear. Here we use data-driven equation discovery to identify the dominant controls on wmax in idealized deep convection. We analyze a set of radiative-convective equilibrium simulations spanning a wide range of sea surface temperatures (290-310 K) and imposed radiative cooling rates (0.75-3 K/day), tracking individual clouds and diagnosing their pre-storm environment and in-cloud properties. Treating the pre-storm environment and in-cloud processes separately, for each we identify a small number of predictors from a broad set of physically motivated variables that robustly explain variations in wmax across simulation regimes. Interpretable equations derived via symbolic regression from the in-cloud variables indicate that latent heating provides the primary acceleration of updrafts, while pressure perturbations act as a leading-order decelerating mechanism that limits peak velocities. Pre-storm predictors such as CAPE and triggering strength constrain the range of possible wmax values, whereas in-cloud condensate loading and pressure effects determine the realized extremes. This work provides a physically interpretable framework for understanding convective updraft intensity, using data-driven analysis informed by existing physical knowledge.
How to cite: Pechacova, A., Casallas, A., Agasthya, L., Beucler, T., and Muller, C.: Data-driven equation discovery of the maximum vertical velocity in idealized deep convection, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7300, https://doi.org/10.5194/egusphere-egu26-7300, 2026.
The Congo Basin is one of the world’s three major tropical convective hotspots, yet it receives substantially less rainfall than the Amazon and the Maritime Continent, despite exhibiting the highest lightning activity globally. This paradox points to fundamental differences in convective structure and rainfall efficiency. Here, we investigate the vertical structure of convection over Central Africa using upward mass transport as a proxy, and examine its relationship with precipitable water and moist static energy (MSE). We show that during the rainy season, convection over the Congo Basin is characterized by a bottom-heavy vertical mass flux profile, accompanied by strong moisture advection in the lower to mid-troposphere. This structure contrasts sharply with the deeper, more top-heavy convective profiles observed over the Amazon and the Maritime Continent, indicating a predominance of relatively shallow convective systems.
Analysis of the MSE distribution reveals high values near the surface, reflecting substantial energy available for convective initiation, while lower MSE aloft is consistent with latent heat release and buoyancy generation within ascending parcels. Together, these results suggest that although the thermodynamic environment over the Congo Basin is highly favorable for triggering convection, the vertical redistribution of mass and energy acts to constrain convective depth and suppress rainfall efficiency. These processes are poorly represented, or entirely missing, in historical CMIP6 Earth system models. Our findings therefore highlight a distinct convective regime over the Congo Basin, with important implications for understanding tropical precipitation and for improving its representation in climate models.
How to cite: Longandjo, G.-N.: The Congo basin as an outlier in the tropical upward mass transport, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20769, https://doi.org/10.5194/egusphere-egu26-20769, 2026.
Understanding the structural and dynamical properties of convective cores is essential to advancing our knowledge of deep convection. Convective cores are the primary engines for the vertical transport of heat and moisture, yet their small spatial scales and the scarcity of vertical velocity measurements make them difficult to observe and represent in numerical weather prediction and climate models.
This study aims to characterize the morphology and intensity of updrafts using a comprehensive dataset of more than 50 Meso-NH km-scale simulations of deep convection events that occurred during five airborne campaigns during which the RASTA (RAdar SysTem Airborne) radar was deployed: CADDIWA, EXAEDRE, HAIC Cayenne, HAIC Darwin, and MAESTRO. These high-resolution simulations encompass a wide spectrum of meteorological environments. We employ an advanced three-dimensional object detection algorithm to isolate convective cores. This volumetric approach allows us to capture the complex geometry of updrafts and the internal variability in vertical velocity.
We specifically investigate the dependence of updraft size and intensity on key environmental parameters using an object-oriented approach. We statistically analyze the distribution of updraft morphological and dynamical properties. Our results show that vertical extension is a driver of updraft intensity, with taller convective cores exhibiting higher vertical velocities. In contrast, the horizontal width of the cores has a significantly smaller impact on their peak intensity. The consequences of these findings for the development of convective parameterizations and for future satellite missions, such as C2OMODO (Convective Core Observations through MicrOwave Derivatives in the trOpics), will be discussed.
How to cite: Mequignon, F., Chaboureau, J.-P., and Richard, J.: Diversity of Convective Updrafts during 5 Airborne Campaigns: Insights from a Large Dataset of km-Scale Simulations, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11105, https://doi.org/10.5194/egusphere-egu26-11105, 2026.
Titan, the largest moon of Saturn, is four times smaller than Earth but has a slightly higher surface pressure (1.5 bar), due to its very extended atmosphere. Titan's atmosphere is mainly composed of nitrogen, with a few percents of methane. The pressure and temperature (~94 K at the surface) conditions there enable a full "hydrological" cycle of methane, with lakes stable on the surface, evaporation, clouds, rain, and rivers.
In this work, we focus on the methane clouds in Titan's troposphere, and in particular on the convective ones.
Titan's methane clouds are monitored from Earth-based telescopes since the end on the 1990's, and have been observed from close by the Cassini spacecraft during half a Titan year (2004 - 2017). Some of the methane clouds in Titan's troposphere are thought to exhibit convective dynamics (e.g. Griffith 2005, 2009, Schaller 2009, Lemmon 2019, Rannou 2021). To understand the processes driving the clouds formation and evolution, modeling is also used, from global scale to regional scale models.
Here we use a regional scale model (160x160 km), with kilometric horizontal resolution. The model is a coupling between the physics of the Titan Planetary Climate Model (Titan PCM, de Batz de Trenquelléon et al. 2025 a,b) and the dynamics of the Weather Research and Forecast model (WRFv4, Skamarock et al. 2019).
We perform idealized simulations with several initial perturbations, at several seasons and locations, in order to constrain in which conditions methane convection appears on Titan.
We obtain condensation for a diversity of setups, with in some cases the triggering of deep convection. We discuss the environments where we find deep convection in our model, and compare them to what has been observed on Titan. We also study convective ascents and the convective available potential energy (CAPE) obtained in our simulations, and compare it to Earth's storms.
References
de Batz de Trenquelléon et al. a, 2025. The New Titan Planetary Climate Model. I. Seasonal Variations of the Thermal Structure and Circulation in the Stratosphere. Planet. Sci. J. 6, 78. https://doi.org/10.3847/PSJ/adbbe7
de Batz de Trenquelléon et al. b, 2025. The New Titan Planetary Climate Model. II. Titan’s Haze and Cloud Cycles. Planet. Sci. J. 6, 79. https://doi.org/10.3847/PSJ/adbb6c
Griffith et al. 2005. The Evolution of Titan’s Mid-Latitude Clouds. Science 310, 474–477. https://doi.org/10.1126/science.1117702
Griffith et al. 2009. CHARACTERIZATION OF CLOUDS IN TITAN’S TROPICAL ATMOSPHERE. ApJ 702, L105. https://doi.org/10.1088/0004-637X/702/2/L105
Lemmon et al. 2019. Large-scale, sub-tropical cloud activity near Titan’s 1995 equinox. Icarus 331, 1–14. https://doi.org/10.1016/j.icarus.2019.03.042
Rannou et al. 2021. Convection behind the Humidification of Titan’s Stratosphere. ApJ 922, 239. https://doi.org/10.3847/1538-4357/ac2904
Schaller, E.L., Roe, H.G., Schneider, T., Brown, M.E., 2009. Storms in the tropics of Titan. Nature 460, 873–875. https://doi.org/10.1038/nature08193
Skamarock et al., 2019. A description of the advanced research WRF version 4. NCAR tech. note ncar/tn-556+ str 145.
How to cite: Moisan, E., Spiga, A., and Chatain, A.: Modeling convective methane clouds on Titan with a kilometer-scale regional model, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13288, https://doi.org/10.5194/egusphere-egu26-13288, 2026.
Elevated moist layers and mid-level clouds around the triple point temperature lead to enhanced radiative cooling at their upper boundaries. This cooling propagates to surrounding areas and creates stable layers that act as buoyancy barriers for neighboring convection and thereby promote the accumulation of water vapor and clouds. Different theories for congestus cloud formation imply different entry points into this feedback loop.
Here, we test the clear-sky convergence (CSC) mechanism -- which predicts a natural maximum horizontal CSC near the triple point caused by the structure of radiative cooling that is determined by the optical properties of water vapor -- using dropsonde measurements from the ORCESTRA field campaign in the Tropical Atlantic. We find that the vertical profile of CSC aligns well with the height of mid-level clouds in the ORCESTRA-East domain, but not in the West. The good agreement in the East is mostly caused by contributions from the stability structure and not from gradients in the radiative cooling.
Additionally, we show using idealized experiments that for a moist adiabatic temperature structure a W-shaped relative humidity is necessary to achieve a positive CSC above the freezing level. This effect is caused by longwave radiative cooling and partly counteracted by shortwave radiative heating. Overall, we conclude that the cooling rate contribution to the CSC mechanism is not enough to trigger a congestus circulation, but can contribute to its maintenance in clear-sky regions with elevated moist layers.
How to cite: Gloeckner, H. M., Schmidt, H., and Stevens, B.: Testing the clear-sky convergence mechanism for congestus cloud formation using ORCESTRA data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17828, https://doi.org/10.5194/egusphere-egu26-17828, 2026.
Tropical convective clouds are a critical component of the Earth system. As these As rain falls from these clouds, it evaporates below cloud base and produces convective cold pools. Cold pools are an important component of the atmospheric system, as they influence surface fluxes, impact the spatial distribution of aerosol, including ice nucleating particles (INP), and can initiate new convection. During summer 2025, as part of the Freshwater Fluxes over the Ocean I – Evaporative Fluxes (FreshOcean) deployment, the RAM-CINC (Relating Atlantic Marine Convection, Ice Nuclei and Cold pools) campaign deployed uncrewed aerial systems (UAS; also known as small drones) aboard the R/V Meteor in the tropical eastern Atlantic Ocean. In RAM-CINC, we characterized properties around convective cold pools, including INPs, bioaerosols, and the thermodynamic environment under quiescent and cold-pool-modified conditions. RAM-CINC’s observations are unique and able to elucidate the complex relationship between the near-surface and marine boundary layer in the small convective cold pool features.
In this presentation, we will give an overview of our field measurements, including novel above-the-surface in situ measurements of several tropical convective cold pools that vary in strength and lifecycle stage. Because our drone measurements were targeted observations before and after cold pool passage, we will demonstrate the impacts of convective cold pool passage both at ship level and above the surface layer. Initial findings indicate maximum changes in temperature of the cold pools from near zero to -1.2 K. Further, we will show aerosol measurements, including characterization of DNA for bioaerosol particles, aerosol concentrations and size distributions, and INPs, from both the drone flights, the surface of the ship, the ocean surface layer, and rainwater, elucidating the link between the near-surface and broader marine boundary layer.
How to cite: Freeman, S., Grant, L., Falk, N., Neumaier, C., Britton, K., Nieto-Caballero, M., Perkins, R., van den Heever, S., Jaeger, L., Ayim, S., Rauch, C., Wurl, O., Thölen, C., Bergfeld, L., Berghöfer, M.-B., Gatti, L., Monroy, D., Haerter, J., and Horstmann, J.: The RAM-CINC Field Campaign: Drone Measurements of Thermodynamics and Ice Nucleating Particles in the Tropical Atlantic’s Cold Pools, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15723, https://doi.org/10.5194/egusphere-egu26-15723, 2026.
Convective aggregation, which often leads to organized convective systems, is ubiquitous over tropical oceans and contributes substantially to both total and extreme precipitation in these regions. Cold pools, crucial processes linked to convection, are known to exert both suppressing and triggering effects on convective initiation, but their role in convective aggregation remains debated. As inspired by the previous study of Biagioli and Tompkins (2023), herea simple two-layer stochastic dynamic model for convective aggregation (see the figure below) is developed by coupling boundary-layer cold pool dynamics with free-troposphere moisture dynamics. The novelty of this model lies in two key aspects: the incorporation of compensating subsidence as a dynamic mechanism to maintain the total number of individual convective cells, and the coupling of the boundary-layer moist static energy (MSE) budget to represent both suppressing and triggering effects of cold pools. The results show that the suppressing effect of cold pools is essential for keeping individual convective cells separated, consistent with observations, while their triggering effect promotes larger cluster sizes and shorter aggregation timescales. The model is then applied to investigate how convective aggregation responds to changes in mean convection lifetime and mean boundary-layer MSE under warming. The parameter sensitivity experiments confirm the robustness of the results and provide insight into phase transitions across different parameter regimes. The model is expected to serve as a theoretical tool for investigating the impact of various physical processes on convective aggregation and may potentially serve as a prototype for convection parameterization that incorporates cold pools and convective organization.
How to cite: Yang, Q. and Li, Y.: The Role of Cold Pools in Convective Aggregation over Tropical Oceans: A Two-Layer Stochastic Dynamic Model, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7045, https://doi.org/10.5194/egusphere-egu26-7045, 2026.
Self-aggregation of deep moist convection has been widely studied in idealized radiative-convective equilibrium. While its relevance to the real tropical climate remains debated, one potential link is zonal convective aggregation within tropical rain belt. However, the mechanisms controlling zonal aggregation are still not well understood. Here, to investigate how convection interacts with the large-scale environment in a zonally symmetric setting, we conduct a series of idealized cloud-resolving simulations in which a meridionally varying, sinusoidal sea surface temperature (SST) distribution is systematically controlled. In particular, we vary the SST amplitude and SST maximum value.
We find that the system selects either a zonally uniform or a zonally aggregated state depending on the following SST parameters: zonal aggregation occurs when both the SST amplitude and the SST maximum are large. We explain this behavior as follows.
For large SST amplitude but low SST maximum, a narrow convergence zone can be maintained by the strong meridional pressure gradient and the associated circulation. In contrast, for large SST amplitude and high SST maximum, the narrow convergence zone cannot be sustained because the circulation weakens as free-tropospheric static stability increases, consistent with convection sitting on a warmer (and more stable) moist adiabat by warmer SST. As a result, convection over the high-SST region cannot maintain surface convergence solely via the meridional overturning circulation driven by subsidence over the low-SST region. Instead, it selects a zonally aggregated state that also extends the circulation in the zonal direction.
How to cite: Yanase, T. and Hohenegger, C.: Controls of Zonal Convective Self-Aggregation in an Idealized Tropical Rain Belt, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9423, https://doi.org/10.5194/egusphere-egu26-9423, 2026.
We show how in the Sapphire configuration of the Icosahedral Non-hydrostatic Model (ICON) the representation of Convectively Coupled Equatorial Waves (CCEWs) is sensitive to the fall speeds of rain and ice, and how reducing these fall speeds can lead to a better representation of CCEWs in ICON. In particular, reductions in the fall speeds of rain and ice lead to more active convectively coupled Kelvin, Inertio-Gravity, and Mixed Rossby-gravity waves and, at the same time, less active convectively coupled Equatorial Rossby waves and Tropical Depressions.
We then explore how changes in these fall speeds upscale from the kilometer scales to the synoptic and planetary scales of CCEW, finding that this up-scaling is mediated by the frequency of occurrence of shallow, stratiform, and deep convection. Reducing the fall speed of rain and ice leads to increases in the frequency of occurrence of, respectively, shallow and stratiform convection profiles and, at the same time, leads to decreases in the frequency of occurrence of deep convection profiles. We argue that changes in these profiles are reflected in the ability of the model to develop the tilt and top-heaviness of CCEWs, which ultimately leads to their better representation.
Our findings suggest a physical representation of CCEWs within ICON and provide further support for the classification of CCEWs into two distinct categories, a gravity wave group and a moisture mode group, each associated with distinct convective profiles and with distinct propagation mechanisms.
How to cite: Ortega, S., Segura, H., Mayta, V., Fiévet, R., Peinado, A., Lee, J., Giorgetta, M., and Stevens, B.: Convectively Coupled Equatorial Waves in ICON, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17407, https://doi.org/10.5194/egusphere-egu26-17407, 2026.
Understanding the life cycle of cumulus clouds remains challenging due to limited knowledge of the factors governing initiation and growth processes. Cloud-cloud interactions—specifically how neighboring clouds influence each other—are a critical but underexplored aspect of evolution of cloud populations. Our key innovation lies in, using both realistic and idealized large-eddy simulations (LES), to identify and quantify competitive relationships among neighbouring clouds as a fundamental driver of cloud population dynamics.
Realistic LES over land reveals a statistically significant pattern: growing clouds suppress the development of their immediate neighbors, suggesting competition for moisture among neighbouring cloudy updrafts. Idealized LES further uncovers that this competition is strongest during the decaying stage of clouds. During this stage, cloud-cloud interactions reduce cloud depths but expand their horizontal extent through environmental moistening—a feedback that effectively reduces cloud spacing and intensifies competition. We also systematically establish the controlling factors of competitive strength by conducting multiple idealized LES simulations by varying: aerosol loading, inter-cloud distance, and surface forcing heterogeneity.
This work establishes cloud-cloud competition as a previously missing process in cumulus evolution theory. By mechanistically resolving how interactions redistribute moisture and energy, we provide a framework to understand population-scale organization. Our findings offer direct pathways to improve cumulus parameterizations in Earth system models, particularly in representing convective clustering and cloud lifetime effects.
How to cite: Chen, J., Xu, C., Lu, C., Hagos, S., Xiao, H., Feng, Z., and Fast, J.: Cloud-Cloud Interactions and Their Roles in the Development of Convective Cloud Systems, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6472, https://doi.org/10.5194/egusphere-egu26-6472, 2026.
Deep convective systems (DCSs) play a crucial role in the tropical hydrological cycle and radiative budget [1,2]. In particular, the largest and longest-lived of those cloud systems contribute to a high fraction of the extreme precipitation in the Tropics [3] Therefore understanding what drives these types of systems is crucial.
To that end, Abramian et al. 2025 [4] developed a new method to predict the maximum area of DCSs using the DYAMOND-Summer simulation with the cloud-resolving global model SAM, and the TOOCAN algorithm to track cloud systems [5]. The method uses simple machine learning models, trained on information on the early stage of the systems and their surrounding environment, including dynamical and thermodynamical variables, morphological features of the systems and the characteristics of their neighbors.
We investigated whether this method would work in observations too, using DCS tracks identified by TOOCAN in satellite data [6], combined with ERA5 data for the physical variables. For both observations and DYAMOND we used both a Lasso linear regression and two different deep neural networks.
Furthermore we aimed at understanding which physical variables constrain the most the maximum area of the systems, and to that end we used an explainable AI method, the integrated gradients ([7]) to assess which physical variables contributed the most to the model prediction.
Firstly, we managed to achieve good predictivity scores for both the non-linear models and both the datasets and we obtained quite robust results in terms of feature importance, with the pre-storm environmental CAPE and deep shear playing a pivotal positive role to achieve a large maximum area, while the presence of neighboring systems was one of the main negative contributors.
Finally, we tested the ML results by looking at composites of the most important variables in the observational dataset: for example pre-storm CAPE composite showed significantly higher than average values for the largest systems.
[1] Nesbitt, S. W., R. Cifelli, and S. A. Rutledge, 2006: Storm Morphology and Rainfall Characteristics of TRMM Precipitation Features.
[2] Bony, S., Semie, A., Kramer, R. J., Soden, B., Tompkins, A. M., & Emanuel, K. A. (2020). Observed modulation of the tropical radiation budget by deep convective organization and lower-tropospheric stability.
[3] Remy Roca and Thomas Fiolleau. Extreme precipitation in the tropics is closely associated with
long-lived convective systems.
[4] S. Abramian, C. Muller, C. Risi, et al. How key features of early development shape deep
convective systems.
[5] Thomas Fiolleau and Remy Roca. An algorithm for the detection and tracking of tropical
mesoscale convective systems using infrared images from geostationary satellite.
[6] T. Fiolleau and R. Roca. A database of deep convective systems derived from the intercalibrated
meteorological geostationary satellite fleet and the toocan algorithm (2012–2020).
[7] Mukund Sundararajan, Ankur Taly, and Qiqi Yan. 2017. Axiomatic attribution for deep networks.
How to cite: Polesello, A., Casallas, A., Muller, C., Roca, R., and Locatello, F.: Physical mechanisms of deep convective system maximum area in a hierarchy of datasets, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17091, https://doi.org/10.5194/egusphere-egu26-17091, 2026.
The top-of-atmosphere radiative effect of tropical anvil clouds varies with cloud opacity, and can range from substantially negative to largely positive. Recent climate model assessments have found a decrease in the proportion of thick, or opaque, anvil cloud with warming, resulting in a positive climate feedback. However, the mechanism for this change remains obscure.
Lifecycle analysis of deep convective clouds tracked using tobac in the convection-permitting global ICOsahedral Non-hydrostatic model (ICON) shows how anvil area and opacity respond to convection. We find that both properties increase in response to increased convective intensity and convective area, but that their sensitivity to each is not equal. To interpret these results, we independently develop a simple analytical model that links anvil expansion and opacity to convective mass flux (CMF). The model predicts that higher CMF leads to greater anvil expansion, increasing the area of thick anvil cloud. But when anvil opacity also depends on convective intensity, we find a strong, non-linear increase in thick anvil amount in response to increasing CMF, consistent with the response observed in ICON. This implies a strong sensitivity of thick anvil amount to changes in the upper tail of the distribution of CMF and illustrates a possible mechanism by which changes in the distribution of cloud CMF could drive anvil thinning in a warming climate.
How to cite: Ritman, M., Jones, W., Stier, P., Senf, F., and van den Heever, S.: Convective controls on anvil area and thickness in analytical and km-scale models, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21234, https://doi.org/10.5194/egusphere-egu26-21234, 2026.
Global circulation models (GCMs) are too coarse to resolve tropical deep convection and convective aggregation, i.e., the clustering of deep convective cells that leads to the formation of mesoscale convective systems. The need for convective parameterizations leads to high inter-model variability in how convective aggregation responds to surface warming, participating in the uncertainty surrounding cloud feedbacks. Realistic kilometer-scale simulations with Global Storm Resolving Models (GSRMs) have recently been run without the need for convective parameterizations under climate change scenarios. We analyze such simulations from the ICON and IFS models and compare them to observations to evaluate whether tropical convective aggregation changes in a warming world. Using the Iorg aggregation metric, we show that simulated tropical convective aggregation becomes increasingly realistic with enhanced horizontal resolution, and that tropical deep convection becomes more aggregated with surface temperatures under uniform warming and under interannual temperature increases. Because convective aggregation helps to cool down the atmosphere through enhanced clear-sky longwave cooling, this could imply that convective aggregation causes a negative climate feedback. However, long-term climate trends of Iorg are less unequivocal, in both observations and models, and results are very sensitive to the method for Iorg computation. This suggests that process-level studies are needed to better understand what drives convective aggregation in the Tropics.
How to cite: Fons, E., Hohenegger, C., and Bony, S.: Does the aggregation of tropical deep convection cause a negative cloud feedback in global storm resolving models? , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21927, https://doi.org/10.5194/egusphere-egu26-21927, 2026.
Observations show a bimodal frequency distribution in total column vapour (TCV) over tropical oceans, with convective rainfall predominantly produced on the moist side of the frequency minimum between two modal peaks. Here we show a km-scale model of the tropics with explicit convection produces a bimodal TCV distribution, whereas the same model with parameterized convection does not. The parameterized model also fails to realistically confine rainfall to a moist mode. Using concepts from statistical mechanics we relate TCV frequency and tendency, and isolate process contributions to tendency in TCV phase-space. Where bimodality is lacking, we find an incorrect relationship between moisture flux convergence and TCV in environments with little or no rainfall. The resulting lack of a strong gradient in TCV tendency with respect to TCV is inconsistent with that expected to maintain a TCV frequency minimum. Our results demonstrate value in the TCV probability distribution as a process diagnostic for the upscale impacts of convection, and as a test for realism in model moisture-dynamics coupling.
How to cite: Bassford, J., Maybee, B., Marsham, J., and Parker, D. J.: Tropical TCV as a process diagnostic: connecting probability to convective processes in km-scale models via moisture budget statistics , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6857, https://doi.org/10.5194/egusphere-egu26-6857, 2026.
Entrainment and detrainment processes remain one of the biggest challenges in convective cloud dynamics and need to be parameterized in mass-flux-based convection schemes. A common assumption in convection parameterizations is that the fate of a mixture between cloud and environmental air is determined by the "buoyancy sorting" hypothesis, and further that the net effect of entrainment and detrainment is to reduce the vertical momentum in the cloud. In this study, Lagrangian trajectories are investigated to understand the entrainment and detrainment processes in maritime shallow cumulus clouds and to examine hypotheses in convection parameterization schemes. Analysis of vertical momentum in cumulus clouds using Lagrangian trajectories and using a bulk budget approach both indicate that the overall impact of entrainment and detrainment on momentum is to accelerate the cloud updraft, rather than acting as a drag force. Following the trajectories, it is found that the entrained air has larger mean vertical velocity than the detrained air, in contradiction with the typical assumption in the mass-flux based plume models. This finding indicates the necessity for a careful treatment of the dynamical properties in the near cloud environment. Investigating the buoyancy of entraining and detraining trajectories, we find that the widely accepted "buoyancy sorting" hypothesis is not able to correctly describe both entrainment and detrainment processes, regardless of how the cloud objects are defined. Instead, whether a mixed parcel is likely to be entrained into or detrained out of the cloud depends on its vertical acceleration. More specifically, vertically accelerated parcels near cloud edge are more likely to be entrained and vertically decelerated parcels are more likely to be detrained. Thus, the "acceleration sorting" hypothesis is proposed. Decomposition of the vertical momentum budget for the entrained and detrained trajectories shows that it is the pressure gradient acceleration, especially the dynamical pressure gradient acceleration associated with the flow structure and the effective buoyancy, rather than the buoyancy alone, that dominate the "acceleration sorting". Our results suggest that the flow structure of cloud thermals might be a potential candidate responsible for "acceleration sorting" processes during the entrainment and detrainment. These findings provide new insights to the understanding of the physical processes during both entrainment and detrainment.
How to cite: Gu, J.-F., Plant, R., Holloway, C., Clark, P., and Stirling, A.: Understanding the Entrainment and Detrainment Processes in Maritime Shallow Cumulus Clouds using Lagrangian Trajectories: Buoyancy Sorting or Acceleration Sorting?, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11967, https://doi.org/10.5194/egusphere-egu26-11967, 2026.
Convection in the tropics is driven by large-scale instability and the buoyancy of convective plumes, but has a variety of representations in Earth system models (ESM). In model convective schemes, subgrid parameterizations approximate the processes that regulate plume buoyancy, leading to substantial spread in simulated tropical thermodynamic structure. While entrainment is known to strongly influence plume buoyancy, recent work highlights an important role for cloud microphysical processes—particularly mixed-phase water physics and precipitation efficiency—in shaping buoyancy during ascent (Emmenegger et al. 2024). The influence of these microphysical processes on tropical buoyancy, and how they should be constrained in convective schemes, is explored here.
Using a perturbed-parameter ensemble (PPE) of the Community Atmosphere Model version 6 (Eidhammer et al. 2024), together with numerical single-column simulations, we construct process-oriented diagnostics to identify controls on convective sensitivity over the tropical western Pacific. Across the ensemble, variability in the convection sensitivity metric is dominated by deep convective parameters, while shallow convective and microphysical parameters exert weaker but systematic influence, revealing a clear hierarchy of physical controls. This hierarchy provides a useful way to organize the role of cloud processes, but does not explain how these parameter sensitivities arise or interact across scales.
To interpret the ensemble behavior, we use a simple bulk plume framework that isolates the effects of entrainment and key microphysical processes on plume buoyancy. Observations from the US Department of Energy’s Atmospheric Radiation Measurement field campaigns together with theoretical considerations of plume buoyancy are used to derive constraints for model representation of these processes. Together, the PPE diagnostics, observational constraints, and bulk plume framework clarify how microphysical and dynamical processes interact to shape simulated tropical convective instability. This approach provides a physics-informed basis for interpreting parameter sensitivity and improving the representation of convection in coarse resolution ESMs.
How to cite: Emmenegger, T., McCluskey, C., Truesdale, J., Neelin, J. D., and Kuo, Y.-H.: Hierarchies of dynamical and microphysical control on tropical convective instability, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14512, https://doi.org/10.5194/egusphere-egu26-14512, 2026.
Convective Organization through Gravity Waves from a Conceptual Model
Ashly Wilson and Jan O. Haerter
Department of Physics and Astronomy, University of Potsdam, Karl-Liebknecht Str. 24/25,
14476 Potsdam, Germany
Correspondence: Ashly Wilson (ashly.wilson@uni-potsdam.de)
Organized convection plays a crucial role in driving extreme weather events, such as
Thunderstorm clusters and tropical cyclones have far-reaching implications for human lives and
infrastructure. It is known that the global tropical circulation is mainly thermally driven (Lau
& Lim, 1982) and that diabatic heating over Earth’s continents plays a key role in
causing Walker and Hadley type circulations. It has long been postulated that tropical deep convection
might couple to different geophysical flows. In a 2D conceptual model, we here propose a two-way interaction where gravity waves can trigger new convection, whereas convection also releases gravity waves.
In our model, a convective ”kick” (in the form of momentum ) (Bretherton and Piotr Smolarkiewicz
1988) initiates gravity waves, which subsequently interact with one another by linear superposition. When a critical amplitude is exceeded, a new convective “kick” results. The physical motivation of the aforementioned convective “kick” is localized heating from convergence in the planetary boundary layer resulting from the interaction between gravity waves, which can act as a source of convection. This enhanced convection, in turn, generates new oscillations within the otherwise stratified troposphere, perpetuating the feedback cycle. The interplay of these processes is proposed as a mechanism of self-organization of convection. Boussinesq equations in the absence of the Earth’s rotation are used. Convection is modeled as a triggered function (Dirac Delta) (Da Yang, 2021).
By extending these concepts, our model provides a simplified yet insightful framework
to explore the dynamics of convective aggregation. Preliminary results suggest that the nonlinear feedback proposed can give rise to a fully-clustered convective system, similar to that seen in convective self-aggregation. Our approach opens avenues for future investigations into the role of gravity waves in
modulating large-scale atmospheric patterns and extreme weather phenomena.
Keywords: Convective Organization, Convectively Coupled Gravity Waves, Triggered
Convection
Abstract for oral session
How to cite: Wilson, A. and Haerter, J.: Convective Organization through Gravity Waves from a Conceptual Model, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21637, https://doi.org/10.5194/egusphere-egu26-21637, 2026.
Convective precipitation is the primary source of freshwater and groundwater recharge in the Arabian Peninsula (AP), occurring as sporadic, localised events. Understanding the Lagrangian properties of these convective systems and how they are influenced by land surface characteristics, topography, and climate change requires convection-permitting modelling, which can be computationally inefficient in such an arid region. However, we trained three machine learning (ML) models -Random Forest (RF), Extreme Gradient Boosting (XGB), and Deep Learning (DL)- to establish relationships between convective environments and precipitation. These models can be applied to any numerical model output (e.g., CMIP6) to infer the probability of convective precipitation, thereby identifying when convection-permitting simulations should be performed. Constrained by CMIP6 temporal (6h) and spatial (~1°) resolutions, we aggregated IMERG V07 precipitation to 6h and averaged over 1 °. We derived 102 features from ERA5 describing moisture, lift, instability, and location to characterise the atmospheric profile. A profile is labelled convective if the accumulated precipitation exceeds the local climatological median, thereby reducing the problem to a binary classification task. The ML classifiers show high skill in identifying convective environments over the northern AP during cold months (Oct-Apr), with a Heidke Skill Score (HSS) of approximately 0.65. However, over the southern AP during warm months (May-Sep), the HSS values drop to around 0.35. These results support the established finding that convective systems over the AP in cold months are linked to large-scale atmospheric patterns. In contrast, in warm months they are localised and/or orographically influenced. Findings also demonstrate that the ML models learned convective environment patterns across the AP. Furthermore, the dirnual performance of ML models remains comparable (HSS: ~ 0.55), except at 12 UTC (HSS: ~ 0.48), when the convection is relatively localised. The SHapley Additive exPlanations (SHAP) method enables the interpretation of each feature’s contribution to the ML models’ prediction. The top three features identified by SHAP differ across ML models. Equivalent potential temperature at 850 hPa, humidity index, and lightning potential index are most important for RF. XGB emphasizes precipitable water vapour, relative humidity at 1000 hPa, and most unstable CAPE. In DL, precipitable water vapour, relative humidity at 700 hPa, and 2m dew-point temperature are the key contributors. The current warming over the AP is +1.22 °C relative to pre-industrial levels. We aim to analyse periods with comparable warming across 10 CMIP6 historical simulations to evaluate the models' ability to reproduce the spatiotemporal distribution and characteristics of convective environments. To assess the effect of climate change, we will analyse two future periods with changes of +2.22 and +3.22 °C from the SSP5-8.5 scenario. SHAP can help evaluate whether the model‑inferred importance ranking of mechanisms controlling convective precipitation changes between present and future simulations. In-depth analysis is undergoing.
How to cite: Homoudi, A., Rust, H. W., Barfus, K., Bernhofer, C., and Mauder, M.: Convective Environments over the Arabian Peninsula in Current and Future Climate, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1893, https://doi.org/10.5194/egusphere-egu26-1893, 2026.
Convective parameterizations used in atmospheric models to represent the effects of unresolved shallow and deep convection on the large-scale flow are traditionally formulated in a diagnostic manner that assumes an instantaneous adjustment of convection to the resolved-scale environment. Prognostic parameterizations, on the other hand, can represent the time evolution and memory of moist atmospheric convection, leading to more realistic interactions between convection and large-scale atmospheric circulation, especially if far from convective quasi-equilibrium. Several such prognostic formulations for the mass flux have been proposed by relaxing the quasi-equilibrium assumption introduced by Arakawa and Schubert [1], based on assumed relations between the convective mass flux and the convective kinetic energy [2,3]. In this work, we develop a new prognostic formulation for shallow and deep convection with cloud cover and convective velocity both being treated as prognostic variables. Interactions between shallow and deep convection are represented in our formulation due to their differing effects on the large-scale environment, but also due to direct cloud-updraft interactions. In radiative-convective equilibrium (RCE), our model predicts that the cumulus cloud cover is proportional to the radiative cooling rate and that the convective velocity depends only on the relative humidity and the tropospheric depth, in agreement with numerical experiments. In addition, our model predicts that convective available potential energy decreases in RCE with the increase of the radiative cooling rate due to the cloud-updraft interaction. Moreover, we show that the inclusion of the cloud-updraft interaction and the cold pools feedback is required for a realistic representation of the diurnal cycle of shallow and deep convection.
[1] Arakawa, A., & Schubert, W. H. (1974). Interaction of a cumulus cloud ensemble with the large-scale environment, Part I. Journal of Atmospheric Sciences, 31(3), 674-701.
[2] Pan, D. M., & Randall, D. D. (1998). A cumulus parameterization with a prognostic closure. Quarterly Journal of the Royal Meteorological Society, 124(547), 949-981.
[3] Yano, J. I., & Plant, R. (2012). Finite departure from convective quasi‐equilibrium: Periodic cycle and discharge–recharge mechanism. Quarterly Journal of the Royal Meteorological Society, 138(664), 626-637.
How to cite: Vraciu, C. and Plant, R.: A prognostic cumulus parameterization with cloud-updraft interaction, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6921, https://doi.org/10.5194/egusphere-egu26-6921, 2026.
In the tropics, heavy daily accumulated precipitation has been shown to be associated with the most organized convective systems. Over the oceans, a simple scaling of the extreme precipitation explains the changes of the high percentiles with SST at a rate close to that predicted from thermodynamics and Clausius Clapeyron. This study aims at clarifying the sensitivity of the deep convective systems relevant to extreme precipitation with respect to SST and to reconcile the object-oriented and the grid box perspectives. For that purpose, two satellite-based precipitation products (IMERG, and GIRAFE) are used together with SST from OSTIA and the deep convective cloud properties from CACATOES. The CACATOES database, a Level-3 product derived from TOOCAN, is available for the 9-year period (2012–2020) over the entire tropical belt (30°S–30°N) on a 1° daily grid.
Our results show that the increase in extreme precipitation fraction is associated with longer lived, larger and slower propagating DCS as the SST warms from 300K to 302.5K. As a consequence, the residence time of the mean system over extreme precipitation grid boxes increases with SST. While it may qualitatively explain the increased accumulated precipitation extreme, akin to the slowdown trend of landing hurricanes, it remains to be shown whether this hypothesis is up to quantitative analysis.
Preliminary investigations of the scaling of the precipitation fraction with SST nevertheless reveals a product-dependence that needs to be explored further using a larger ensemble of satellite precipitation products. The physical robustness of the hypothesis is also further analysed by looking at the thermodynamical and dynamical environment, using ERA-5, of the Deep Convective System relevant to the extreme precipitation grid boxes. A moistening of the environment, consistent with the precipitation increase is found. The changes in the dynamical environment will be further discussed at the conference.
How to cite: Rajasekharan Sujatha, A. and Roca, R.: Are increases in extreme accumulated precipitation over tropical oceans with SST warming due to the slowing of deep convective systems?, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10223, https://doi.org/10.5194/egusphere-egu26-10223, 2026.
The space-borne C3IEL (Cluster for Cloud evolution, ClImatE and Lightning) mission, developed jointly by the French and the Israeli space agencies, aims at providing new insights on convective clouds, at high spatial and temporal resolutions, close to the scales of the individual convective eddies. The mission will simultaneously characterize the convective cloud dynamics, the interactions of clouds with the surrounding water vapor, and the lightning activity.
The C3IEL mission consists in a short-baseline (~150 km) train of 2 synchronized small satellites. Each satellite carries a visible camera (670 nm) for cloud imagery at a spatial resolution of ~20 meters, near-infrared water vapor imagers (1.04, 1.13 et 1.37 µm ; ~125 m at nadir) measuring in and near the water vapor absorption bands, a lightning imager (777.4 nm ; ~140 m at nadir; 10 ms time resolution) and three photometers (337, 391 and 777.4 nm; 50 μs time resolution).
The scientific objectives of the C3IEL mission will be first reminded. They consist in documenting the convective cloud development through their 3D evolution and their environment with the retrieval of water vapor surroundings the clouds. In addition, the lightning activitiy created by such clouds will be observed. The presentation will first remind the satellite train configuration, the different sensors of the mission and the innovative and different observational strategies that will be applied during daytime and nighttime. We will then provide an update on the expected observations and products.
Figure : Artist view of the C3IEL, Cluster for Cloud evolution, ClImatE and Lightning, mission. © CNES - Olivier Satteler.
How to cite: Cornet, C., Rosenfeld, D., Aviad, S., Cheymol, C., Defer, E., Deschamps, A., Frid, A., Gillot, L., Holodovsky, V., Kaidar, A., Navas, J.-B., Penide, G., Price, C., Ricard, D., Rimboud, A., Schechner, Y., Truffier, A., Yair, Y., and Zemb, A.: C3IEL, the Cluster for Cloud evolution ClImatE and Lightning Mission to Study Convective Clouds at High Spatial and Temporal Resolutions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11929, https://doi.org/10.5194/egusphere-egu26-11929, 2026.
Convective memory describes the extent to which current convective state is dependent on prior states. A linear state space model can predict the evolution of horizontal mean profiles and attempts to capture convective memory in its latent space. These state space models rely on black-box model structures to produce responses, which necessitate additional approaches to interpret causes of convective behavior.
To this end, for a more physically-interpretable approach to the latent state and its dynamics, we use probability distribution function (PDF) or histograms of idealized cloud-resolving model simulations. These PDFs, which contain joint distributions of thermodynamic variables, including temperature and moisture, provide more information than horizontal averages. This representation allows for a more physical picture of the latent state, while still avoiding the complexity of the three-dimensional spatial domain. We can further reduce the data volume by applying dimensional reduction techniques.
By observing the PDF correspondence to temperature and moisture tendencies and other convective effects, we will attempt to use these physical insights to predict time series of horizontal mean profiles, including the nonlinear response of atmospheric perturbations.
How to cite: Shen, W. and Kuang, Z.: Moist convective memory in terms of thermodynamic joint probability distributions , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15814, https://doi.org/10.5194/egusphere-egu26-15814, 2026.
Convective self-aggregation (CSA) is when tropical deep convection self-organizes in radiative-convective equilibrium (RCE) simulations whereby the moisture field spontaneously separates into strongly convecting and dry, subsiding, subregions. Departing from the classical, constant sea surface temperature, RCE setup, a recent study found that including a diurnal cycle to mimic surface temperature variations between night and day over tropical land, e.g. the Sahel region, enables the onset of CSA. In contrast, corresponding simulations with constant surface temperature, which might emulate the atmosphere over the ocean, showed no strong aggregation (Kruse et al., 2025, Jensen et al., 2022, Haerter et al., 2020). Furthermore, once formed, the “diurnal self-aggregation” remained in place, when surface temperature was then set constant, suggesting a hysteresis effect.
We here explore how this diurnal effect is modified by wind shear - a crucial ingredient for realistic Sahelian conditions. We conduct idealized cloud resolving simulations of the tropical atmosphere using the System for Atmospheric Modeling (SAM), version 6.11 (Khairoutdinov and Randall, 2003). The simplified boundary conditions include an irrotational RCE atmosphere with doubly periodic lateral boundaries. We mimic conditions over (i) land by prescribing diurnal sinusoidal surface temperature oscillations and (ii) over the ocean by prescribing a constant sea surface temperature. To mimic wind shear we nudge towards a prescribed wind profile which is based on ERA5 data from tropical northern Africa.
In simulation runs (Kruse, 2024), we observe that the level of aggregation and other variables oscillate over time. With our simulations we investigate how these frequencies relate to the chosen domain. Additionally, we explore whether not only the simulations with a diurnal cycle but also simulations with realistic wind shear and a constant sea surface temperature show CSA after multiple weeks. Our analysis has relevance for the understanding of convective clustering over tropical land and the persistence of such clusters when advected over the tropical ocean - thus harboring conclusions for tropical cyclogenesis.
Haerter, Jan O., Bettina Meyer, and Silas Boye Nissen. "Diurnal self-aggregation." NPJ Climate and Atmospheric Science 3.1 (2020): 30.
Jensen, Gorm G., Romain Fiévet, and Jan O. Haerter. "The diurnal path to persistent convective self‐aggregation." Journal of Advances in Modeling Earth Systems 14.5 (2022): e2021MS002923.
Khairoutdinov, Marat F., and David A. Randall. “Cloud Resolving Modeling of the ARM Summer 1997 IOP: Model Formulation, Results, Uncertainties, and Sensitivities.” Journal of the Atmospheric Sciences 60, no. 4 (2003): 607–25.
Kruse, Irene L. “Chasing the Storms. A Simulation and Observation-Based Exploration of Mesoscale Convective Systems and Cold Pools, from the Midlatitudes to the Tropics.” PhD Thesis, University of Copenhagen, 2024.
Kruse, Irene L., Romain Fiévet, and Jan O. Haerter. "Tipping to an aggregated state by mesoscale convective systems." Journal of Advances in Modeling Earth Systems 17.3 (2025): e2024MS004369.
How to cite: Bergfeld, L. and Haerter, J. O.: Exploring the hysteresis of tropical diurnal self-aggregation under realistic wind shear conditions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20068, https://doi.org/10.5194/egusphere-egu26-20068, 2026.
Global storm-resolving simulations at kilometer scales (1–5 km) provide new opportunities to represent convective processes, yet they remain in the gray zone of deep convection, where cumulus parameterization choices can strongly affect model performance. Using global kilometer-scale simulations from the Digital Earth Global Hackathon 2025, this study applies a new vortex–mesoscale convective system (MCS) tracking and matching algorithm to examine how two convection parameterization configurations—turning off deep convection (IFS-deepoff) and deep convection with reduced cloud-base mass flux (IFS-rcbmf)—influence the simulation of MCSs, vortices, and precipitation during the 2020 Meiyu season over East Asia. Results show that IFS-deepoff outperforms IFS-rcbmf in reproducing the total amount and spatial distribution of precipitation, although both schemes overestimate MCS frequency and their contribution to rainfall over the Sichuan Basin and the middle–lower Yangtze River. Importantly, precipitation biases are not governed by MCS frequency alone, but depend strongly on the coupling between MCSs and vortices. Precipitation in both schemes is highly sensitive to vortex simulation, with IFS-deepoff producing stronger extremes due to enhanced moisture convergence associated with boundary-layer vortices and increased convective available potential energy (CAPE). These findings highlight vortex–MCS coupling as a critical control on precipitation in the kilometer-scale gray zone, demonstrating that convection parameterization influences rainfall primarily through its modulation of multiscale dynamical interactions. This study provides new insight for improving convection treatment in next-generation global storm-resolving models.
How to cite: Xi, X., Zhang, Y., and Sun, J.: Vortex–MCS–Precipitation Linkages: Sensitivity to Cumulus Convection Parameterization in Global Kilometer-Scale Models during China’s Meiyu Season, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20742, https://doi.org/10.5194/egusphere-egu26-20742, 2026.
This study investigates the response of convectively coupled Kelvin waves (CCKWs) under two warming scenarios using the ICOsahedral Non‐hydrostatic model (ICON) in its global storm-resolving configuration. A control simulation (CTRL) is conducted using prescribed historical (1979–1997) sea surface temperatures (SSTs). Taking CTRL as reference, a simulation with a homogenously SST increase of 4 K (+4K) and another with a 4-fold increase in atmospheric CO2 concentration (4×CO2) are conducted. Results show that CCKWs are substantially strengthened in the +4K experiment, exhibiting enhanced spectral power, faster phase speeds, and more active spatial activity, whereas nearly no change is found in the 4×CO2 experiment. Under uniform SST warming, enhanced surface energy input-dominated by increased latent heat flux-supports stronger and deeper tropical convection, increasing the strength and coherence of convection-circulation coupling. We hypothesize that this enhanced coherence may shorten the adjustment timescale between convective heating and wave-related circulation, resulting in faster eastward propagation and enhanced wave variability. These results highlight the critical role of surface-driven flux enhancement in modulating convectively coupled tropical wave activity under warming.
How to cite: Ji, X., Segura Cajachagua, H. M., Ortega Arango, S., and Feng, T.: Response of convectively coupled Kelvin waves under warming scenarios in a global storm-resolving model, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3016, https://doi.org/10.5194/egusphere-egu26-3016, 2026.
Convective storms can develop rapidly, creating hazards to local populations via intense precipitation, strong winds and lightning. The large-scale environment in which thunderstorms develop is often well-captured in forecast systems yet predicting where individual storms will initiate remains a fundamental challenge. It is known that differential heating driven by soil moisture patterns creates atmospheric circulations which favour convective initiation over drier soils, whilst wind shear between low and mid-levels can enhance upscale storm growth.
Here we show that the most explosive initiations are especially favoured over soil moisture contrasts via an interaction with wind shear. Analysing 2.2 million afternoon convective initiations across Sub-Saharan Africa identified from Meteosat Second Generation imagery for the period 2004-2023, we find that stronger low-level directional wind shear systematically enhances the sensitivity of convective initiations to underlying soil moisture gradients (as identified by combining ERA5 wind fields, MSG land surface temperature and Advanced Scatterometer soil moisture). We detect 68% more initiations classed as extreme given favourable (versus unfavourable) soil conditions, with the most rapidly deepening clouds occurring where soil moisture-induced circulations oppose the direction of mid-level cloud displacement. We propose this configuration promotes wider, more resilient updraughts capable of overcoming shear-enhanced entrainment. Furthermore, where mid-level wind opposes the low-level flow, we find subsequent rainfall to be strongly correlated with locally drier soils as developing rainy clouds follow the mid-level wind direction. Whilst such shear conditions are particularly common over Tropical North Africa, the effect favours negative soil moisture-precipitation feedbacks globally. The combination of soil moisture heterogeneity and wind shear provides a potentially important source of predictability for where deep convection develops, particularly for the most rapidly-developing thunderstorms.
How to cite: Klein, C., Taylor, C., Barton, E., Hahn, S., and Wagner, W.: Wind shear enhances soil moisture influence on rapid thunderstorm growth, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5873, https://doi.org/10.5194/egusphere-egu26-5873, 2026.
Using the aerosol--aware Weather Research and Forecasting (WRF) model in an idealized tropical island framework, we examine how boundary layer moisture modulates the convective response to cloud condensation nuclei (CCN). Simulations across CCN concentrations of 30--2400 per cubic cm and varying relative humidity, CAPE and surface–flux regimes, show that convection consistently weakens with increasing CCN, but only when the boundary layer is sufficiently moist. In such a setting, fewer CCN yield larger droplets that rapidly convert to rain and reduce evaporation at mid-levels (between 2 and 4km), both of which warm and dry the layer and thereby weakening shallow convection. This limits vertical transport of moist static energy (MSE), allowing near-surface MSE and Convective Available Potential Energy (CAPE) to build up. As a result, subsequent deep convection in clean cases exhibits stronger updrafts, greater graupel production, and enhanced convective and mass fluxes. In contrast, humid but polluted environments yield numerous small drops which remain lofted and suppress warm rain, enhancing the evaporative source of vapor at mid-levels, strengthening shallow convection, limiting CAPE growth, and ultimately producing weaker convection.However, when the boundary layer is dry, both CAPE and convection show little sensitivity to CCN concentration, highlighting the role of moisture preconditioning. Satellite composites of TRMM convective intensity (measured by 40 dBZ echo top height) and MODIS droplet number (Nd) over tropical islands tentatively appear to support this mechanism. Higher Nd values are associated with lower CAPE and weaker convective vigor, with the strength of the trend being proportional to the near-surface relative humidity - consistent with simulations. Together, these results suggest that in a tropical island-setting, aerosols impact convection primarily when the boudary layer is preconditioned with sufficient moisture, and that under these conditions increased aerosol loading tends to suppress rather than invigorate deep convection.
How to cite: Robinson, F., Storelvmo, T., Sherwood, S., and Kirshbaum, D.: Humidity-Dependent Sensitivity of Tropical Island Deep Convection to Aerosols, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6908, https://doi.org/10.5194/egusphere-egu26-6908, 2026.
This study investigates the morphological structure, propagation, and evolution of Mesoscale Convective Systems (MCSs) associated with four historical extreme rainstorm events in Ningxia (August 21, 2016; July 22, 2018; August 9, 2022; and August 24, 2024). The analysis utilizes high-resolution observational data, including the China Severe Weather Automatic Nowcasting(SWAN) radar mosaic, three local C-band radars, and 1020 regional automatic weather stations. The results classify these extreme precipitation MCSs into two distinct categories: (1) Topography-dominated Back-Building/Quasi-Stationary (BB/QS) systems characterized by a single rainstorm center (e.g., the "8·21" and "7·22" events); and (2) Composite Multi-MCS types (Training Line/Adjoining Stratiform, TL/AS and Embedded Lines, EL) characterized by dual rainstorm centers (e.g., the "8·9" and "8·24" events). The BB/QS MCS concentrates heavy rainfall along the eastern foothills of the Helan Mountains. Driven by orographic lifting and cold pool dynamics, new convective cells initiate on the southern flank and propagate northward (back-building), resulting in a quasi-stationary system. These systems exhibit deep convection with 40 dBZ echoes reaching 12 km and the 60 dBZ centroid located around 9 km (above the 0°C level). This "high-centroid, strong ice-phase" structure yields high precipitation efficiency, producing accumulated rainfall exceeding 240 mm. In contrast, the TL/AS MCS affects the eastern banks of the Yellow River. Here, convective cell motion is highly parallel to the orientation of the convective line, leading to a significant "training effect" where cells continuously regenerate upstream and propagate downstream. This mode features a lower convective centroid, with the 60 dBZ center located approximately at 4 km (below the 0°C level), indicating a typical "low-centroid, warm-rain" process that results in accumulations exceeding 200 mm. Furthermore, the EL MCS in the arid northwest region is notably modulated by the dry low-level environment, causing a marked contraction of the stratiform cloud region. Strong echoes (>40 dBZ) are generally confined below 6 km. Due to weaker updrafts, the EL phase itself produces limited rainfall (about 20 mm); its primary disaster risk stems from Nonlinear (NL) convective systems triggered during the system's evolution. Based on these findings, a conceptual evolution model for extreme rainstorm MCSs in Ningxia is established, providing a theoretical basis for monitoring and early warning of extreme precipitation in the arid regions of Northwest China.
How to cite: Li, A. and Chen, Y.: Evolution Laws and Conceptual Models of MCS for Different Types of Extreme Rainstorms in the Arid Region of Northwest China, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2165, https://doi.org/10.5194/egusphere-egu26-2165, 2026.
The organization of deep convection into squall lines is not represented in global climate models, despite being among the biggest storms on Earth and having a huge impact on precipitation and extremes in tropical regions such as the Sahel or the Amazon [1, 2]. The overarching goal of this project is to parameterize the occurrence of squall lines in the general circulation model LMDZ, and their impact on their environment. Since the interaction between wind shear and cold pools is essential in squall line formation and maintenance [3], we plan to take advantage of the cold pool scheme [4] already implemented in LMDZ.
Regarding the occurrence of squall lines, we hypothesize that it can be predicted based on the domain-mean wind profile and the cold pool properties, diagnosed from the cold pool scheme.
Regarding the impact of squall line in their environment, we hypothesize that squall lines impact the vertical profile of diabatic heating [5], either through entrainment in convective updrafts [6] or through the rate of rain evaporation [7]. We also hypothesize that the larger rate of rain evaporation strengthens cold pools favoring more likely and more intense convection [8]. Entrainment and rate of rain evaporation can be varied in the convective scheme of LMDZ [9] through tunable parameters.
To test these hypotheses, we analyse simulations from the cloud-resolving model SAM in radiative-convective equilibrium configuration with various wind shear and large-scale ascent conditions.
Ultimately, the representation of squall lines should improve the simulation of precipitation distribution, variability and extremes in tropical regions such as the Sahel or the Amazon.
[1] R. Roca, J. Aublanc, P. Chambon, T. Fiolleau, N. Viltard, J. Clim. 27, 4952–4958 (2014).
[2] R. Roca, T. Fiolleau, Com. Earth & Env. 1, 18 (2020).
[3] R. Rotunno, J.B. Klemp, M.L. Weisman, J. Atmos. Sci., 45(3), 463-485 (1988).
[4] J.-Y. Grandpeix, J.-P. Lafore, J. Atmos. Sci., 67, 881–897 (2010).
[5] U. Anber, S. Wang, A. Sobel, J. Atmos. Sci., 71, 2976–2993 (2014).
[6] T. Becker, C. S. Bretherton, C. Hoehenegger, B. Stevens, Geophys. Research Lett. 45, 455–462 (2018).
[7] J. P. Lafore, J. L. Redelsperger, G. J. Jaubert, Atmos. Sci., 45(22), 3483-3500 (1988).
[8] G. G. Rooney, A. J. Stirling, R. A. Stratton, M. Whitall, Quart. J. Royal. Meteoro. Soc. 148, 962–980 (2021).
[9] K. A. Emanuel, J. Atmos. Sci. 48, 2313–2329 (1991).
How to cite: Segueni, M., Risi, C., and Rochetin, N.: Organization of deep convection into squall lines, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5758, https://doi.org/10.5194/egusphere-egu26-5758, 2026.
Marine shallow clouds are highly abundant over subtropical oceans and play a key role in regulating Earth’s radiative balance, exerting a significant net cooling effect. However, their representation in current climate models remains incomplete, contributing substantially to uncertainties in climate projections and cloud feedback mechanisms. Marine shallow clouds have conventionally been classified into closed cells, open cells, and shallow cumulus clouds. Recent advances in deep learning have expanded this classification by enabling the automatic detection of additional pattern types. Yet, even within these pattern classes lies a spectrum of finer patterns that have not been systematically identified, leaving their underlying dynamics and radiative impacts underexplored.
We present an AI architecture for the classification of marine shallow cloud satellite images into newly defined fine pattern classes. We focus specifically on marine shallow cloud regimes with large cloud fraction values (above 0.9), which we partition into four classes: closed cells, closed cloud streets (rolls), stratiform clouds and unorganized convection. This finer partitioning enables the extraction of richer information from large cloud fraction conditions and a more detailed investigation of physical processes governing their organization. The training dataset was labeled by the Weizmann Cloud Physics Group members. Explainable AI tools are used to analyze the model’s internal representations and learn how it differentiates between the new classes. Applying the trained model to a large number of satellite images enables us to construct a novel, comprehensive and systematically classified database of cloud patterns. The availability of this extensive dataset allows the use of remote sensing cloud properties to characterize the unique features of each regime, and radiative flux data to assess their distinct radiative behaviors, despite their similarly high cloud fraction values. This work provides a clearer understanding of marine shallow cloud patterns and offers insight into the relationships between cloud morphology, underlying dynamics, and radiative effects.
How to cite: Roth, H., Dror, T., Sarafian, R., Dekel, G., Altaratz, O., and Koren, I.: Detailed classification of marine shallow clouds with large cloud fraction using artificial intelligence, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6979, https://doi.org/10.5194/egusphere-egu26-6979, 2026.
Changes in tropical precipitation, particularly extremes, are closely linked to mesoscale deep convective systems (DCSs). While previous work has largely focused on precipitation characteristics, much less is known about how the DCSs which produce extreme rainfall respond to warming. Recent studies showed that the Lagrangian perspective offered by storm tracking in satellite imaging was promising. Here, we exploit two models (SAM and MesoNH) of idealised tropical convection from the RCEMIP project, on which the TOOCAN DCS tracking algorithm has been applied, to study DCSs life cycles in an idealised setup. Focusing on their onset rate, lifetime, area and precipitation intensity, we show that despite the relatively small increase in domain-mean precipitation (+2.5 %/K), the characteristics of DCSs change much more with warming, but their respective responses mostly compensate. Mean DCS precipitation intensity increases in both SAM (+10 %/K) and MesoNH (+2.3 %/K). However, the two models predict strong but opposite responses in DCS area and onset rate—increased area (+8.0 %/K) and lower onset rate (–13 %/K) in SAM, the opposite in MesoNH (–4.2 %/K and +7.4 %/K, respectively). This may be related to their different organisation response to warming. Yet, we find that the probability density functions (PDFs) of DCS lifetime, area and precipitation intensity normalised by their respective ensemble average over all DCSs, are climate and model invariant: the PDF of each of these variables is identical in both the colder and the warmer simulation, in both SAM and MesoNH. If confirmed, such a climate invariance could fuel further research about the physical mechanisms of extreme storms response to warming.
How to cite: Clavier, N., Bao, J., and Muller, C.: Invariance in convective storms with warming: A Lagrangian view, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7943, https://doi.org/10.5194/egusphere-egu26-7943, 2026.
The impacts of the anthropogenic heat (AH) effect on the evolution of a merger-formation bow echo over the Guangdong-Hong Kong-Macao Greater Bay Area are documented. The utilization of radar data assimilation greatly improves the simulated results comparing against observations, strengthening the robustness of analyses in this work. The simulation with AH effect produces the most accurate results compared to observations, exhibiting approximately 62% larger spatial extent of heavy rainfall (> 30 mm) and twice the area of strong winds (> 10.8 m s-1) compared to the non-AH simulation. Additionally, the top 1% rain rates and surface winds from the AH-included simulation are about 25% stronger and 23% greater, respectively, relative to the non-AH counterpart. On the one hand, higher AH flux tends to enhance the values of CAPE and vertical wind shear within urban areas on average, providing favorable thermodynamic environmental conditions for convective development. On the other hand, greater AH effect triggers stronger convective cell, leading to more intense merged system. This cell plays a crucial role in the merger process and the formation of bow echo, but it does not persist sufficiently in the non-AH simulation. A third sensitivity simulation, excluding the urban land cover, produces results comparable to those of the non-AH simulation. This study quantifies the relative contribution of the AH effect to the evolution of convective systems and the associated weather-related hazards over the Greater Bay Area, underscoring the significant impacts of AH forcing on the regional flow patterns and the corresponding convection dynamics.
How to cite: Zhou, A. and Zhao, K.: The Impact of Anthropogenic Heat Effect on the Evolution of a Merge - Formation Bow Echo in the Greater Bay Area of China , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8801, https://doi.org/10.5194/egusphere-egu26-8801, 2026.
Measuring vertical velocity is crucial for advancing our understanding of deep convection. The Convective Core Observations through MicrOwave Derivatives in the trOpics (C2OMODO) mission aims to retrieve vertical velocity using ice-sensitive microwave measurements taken at two closely spaced time intervals. In preparation for the mission, it is essential to investigate the relationship between vertical velocity and satellite measurements. However, vertical velocity within convective cores is rarely measured, and no comprehensive dataset currently exists. To address this, it is necessary to create datasets that combine vertical velocity with corresponding synthetic satellite observations. The recent porting of the Meso-NH non-hydrostatic mesoscale model to GPU architectures enables the efficient generation of such high-resolution datasets. We present the RASTA (RAdar SysTem Airborne) collection, a dataset of Meso-NH kilometer-scale simulations developed for the C2OMODO project. It is based on five airborne field campaigns during which the RASTA radar was deployed: CADDIWA, EXAEDRE, HAIC Cayenne, HAIC Darwin and MAESTRO. The dataset comprises 13 billion atmospheric columns from 53 simulations, validated against RASTA radar data and satellite observations. The exploration of the convective cores variability in the dataset is carried out using formal concept analysis (FCA), a mathematical framework that links objects and attributes through a Galois connection. FCA is used to classify convective cores according to the environmental factors that most influence them. The resulting concept lattices reveal which combinations of conditions favor convection. They also highlight common patterns and specific characteristics in different meteorological environments.
How to cite: Richard, J. and Chaboureau, J.-P.: Variability of Convective and Ice Cloud Structures in the RASTA Collection of Kilometer-Scale Meso-NH Simulations for C2OMODO, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11838, https://doi.org/10.5194/egusphere-egu26-11838, 2026.
In the tropics, the spatio-temporal aggregation of deep convective systems (DCS) has strong implications for the regional and global energy balance, for cloud-circulation interactions, and for the production of heavy precipitation on various scales. Yet, no ubiquitous definition or tracking algorithm exists for identifying cloud "aggregates", or cloud "clusters" in an objective way. Current cloud tracking tools are either developed for detecting individual cloud entities, or for identifying cloud clusters faithful to the original definition of mesoscale convective systems. They thus exclude a wide variety of DCS ensembles that organize in space and time without always merging into contiguous structures.
This work introduces a methodology to define and track coherent aggregates of deep convective systems using infrared brightness temperature (Tb) retrievals from geostationary satellites intercalibrated across the tropical band. The novelty lies in the combination between a spatiotemporal clustering and a region-growing method: starting with the coldest and largest cloud cores, we gradually attribute to preexisting aggregates the new neighboring systems that appear when the Tb threshold is increased to warmer values, until reaching the warmer outer edge of anvil clouds. Using a few criteria to measure organizational properties of the segmentation, we tune the algorithm parameters to maximize the realism of final cloud aggregates, minimize split-and-merge issues, and ensure that each aggregate is not intertwined with its neighbors whenever possible.
We demonstrate the good performance of this algorithm through a variety of realistic modes for deep convective organisation. The example case studies chosen include mesoscale convective complexes and cyclones over tropical oceans, cloud clusters embedded in African Easterly Waves, and the upscale merging of DCS in the course of the diurnal cycle of convection over tropical continents. A few composite properties of aggregates are shown for the entire tropics, to illustrate the potential of this dataset in providing new insights on the variety of organized convective patterns and on their associated multiscale interactions. The dynamics of individual systems can now be explored as an integral component of larger convective aggregates, across the diversity of cluster morphologies that populate the current tropics.
How to cite: Fildier, B. and Saba, O.: Tracking aggregates of deep convective systems in the tropics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12752, https://doi.org/10.5194/egusphere-egu26-12752, 2026.
Rising thermals are a fundamental component of atmospheric convection. Representing small parcels of air with relatively low density compared to their environment, thermals are known to carry a significant part of the vertical mixing in convective layers. Coherent structures in various modes of convection have been observed to consist of sets of thermals, in the form of vertical chains but also exhibiting more complex spatial structures. The recent focus on the spatial organization of convection has renewed interest in this topic, with meteorological field experiments providing new insights into thermal behavior. However, key aspects of thermal life cycle and population statistics remain unknown. To fill this data gap, this study tracks thermals in large-eddy simulations of a diurnal cycle of convection over land. Based on measurements on 30 August 2016 during the Hi-SCALE field campaign at the ARM SGP site, the case features a shallow convective boundary layer transitioning into deep convection during the late afternoon. Adopting previously proposed tracking algorithms, hundreds of thousands of thermals are thus identified and analyzed. Apart from investigating thermal life-cycle statistics and their diurnal evolution, a key research objective is to gain insight into what controls the birth rate of such thermals. Good scaling of thermal birth rates with various integrated buoyancy scales is reported, distinguishing between various layers and various thermal classes. The birth rate of dry thermals in the sub-cloud layer scales well with the surface-driven Deardorff convective velocity scale. Thermal presence in the cloud layer is found to be partially driven by local buoyancy scales, but is significantly boosted by thermals rising into the cloud layer during the shallow convective phase. This surface coupling disappears during the transition to deeper convection in the late afternoon, after which thermal birth rates are generally lower but scale well with the cloud layer buoyancy flux. The implications and potential use of these results for the conceptual modeling of convective organization and its representation in larger-scale circulation models are briefly discussed.
How to cite: Neggers, R. and Bartolomé García, I.: On the birth rate of thermals in convective layers, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14071, https://doi.org/10.5194/egusphere-egu26-14071, 2026.
Operational forecasts of monsoon heavy rainfall suffer from the inadequate representation of convective lifecycle and three-dimensional organization of moisture. This study introduces a novel trajectory-based lifecycle framework that synchronizes atmospheric profiles with precipitation stages for dynamic tracking of convective. Using 16 years (2005–2020) of ERA5 reanalysis and rain-gauge data over the subtropical monsoon region, Hong Kong, we diagnose the coevolution of Integrated Relative Humidity (IRH), Moist Static Energy (MSE), and dynamical fields across rainfall intensities. Results show that rainfall intensity correlates with the moist layer depth: organized deep convection (e.g., heavy rain) is coupled with full-tropospheric saturation (IRH ≥0.88 and ≥0.85 at the lower- and mid-upper tropospheric layers) and robust ascent, whereas disorganized shallow convection (e.g., light rain) is confined to the lower troposphere. Critically, the framework reveals an IRH trajectory of an upward-expanding moistening pattern synchronized with peaking MSE gradients (>10 kJ kg⁻¹) and strengthening low-level convergence, underscoring a coupled energy-dynamic sequence fundamental to the lifecycle of organized convective systems. Moreover, IRH anomalies from seasonal baselines are more consistent predictors of intense events than absolute thresholds, highlighting the importance of environmental preconditioning. This trajectory-based approach provides physical insights into convective organization in subtropical monsoons and a process-oriented tool for evaluating and improving the representation of convection in models.
How to cite: Wan, M., Su, H., and Chan, P. W.: A Trajectory-Based Framework for Diagnosing Convective Lifecycle and Organization in Monsoon Rainfall: Coupled Evolution of Moisture, Energy, and Dynamics , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15571, https://doi.org/10.5194/egusphere-egu26-15571, 2026.
The representation of deep moist convection and organized convective systems in numerical modeling of the atmosphere is pivotal to model fidelity. Although kilometer-scale global models better capture large-scale oscillations and convection features that traditional general circulation models struggle to represent, past studies have reported notable differences in their simulations of convective organization, especially over the tropics. However, untangling the controlling factors of convective organization remains challenging, since physical processes associated with deep moist convection are intrinsically multiscale and deeply intertwined. To address this challenge, this study develops a cumulus parameterization tailored for kilometer-scale models, aiming to enable the modulation and systematic testing of multiscale interactions associated with deep moist convection.
The cumulus parameterization developed in this study employs Arakawa's unified parameterization to represent the interactions of unresolved deep moist convection with its environmental flow, including the explicitly simulated convection. The underlying parameterizability of unresolved deep moist convection follows the notion of convective quasi-equilibrium, while a complementary closure is employed to predict the fractional area covered by cumulus updrafts and adjust the local vertical eddy transports accordingly. A representation of stochasticity and convection memory is introduced by coupling our parameterization with a cellular automaton. Empirical values and physical assumptions are used to establish our parameterization as a prototype subject to designing systematic experiments of specific process representation in the future.
Idealized experiments of tropical maritime deep convection at a horizontal grid spacing of 3 km demonstrate that employing our parameterization in convection-permitting simulations modulates deep convective system features while generally retaining the profile of total vertical energy transport associated with deep convection. Unresolved deep convection dominates the vertical energy transport in the early stages of deep convective systems, and its contribution gradually weakens as systems develop. Meanwhile, detrainment from unresolved deep convection leads to the early occurrence of cumulus congestus and cumulonimbus. In comparison with the ordinary convection-permitting simulation, our parameterization exhibits a pronounced congestus mode in the vertical velocity profile of mature convective systems, partially due to the enhanced dry static stability. Overall, simulations with our parameterization exhibit fewer and larger short-lived convective systems. Further investigation into the source of the uncertainty in convective organization is warranted.
How to cite: Su, C.-Y. and Peters, J.: Enabling Systematic Modulation of Deep Convective Systems in Kilometer-scale Models using a Unified Cumulus Parameterization, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15604, https://doi.org/10.5194/egusphere-egu26-15604, 2026.
NASA’s Investigation of Convective Updrafts (INCUS) aims to improve understanding of how, when, and why tropical convective storms form and why only some lead to extreme weather. Much of the vertical transport of water and air between Earth’s surface and the upper troposphere is facilitated by convective storms. This vertical transport of water and air, often referred to as convective mass flux (CMF), plays a critical role in Earth’s weather and climate system through its impacts on precipitation rates, detrainment and upper tropospheric moistening, high cloud feedbacks, and the large-scale circulation. Recent studies have also suggested that CMF may change with changing climates with subsequent implications for flood-producing rainfall, severe weather, and lightning. In spite of the critical role of this vertical transport of water and air within the weather and climate system, much is still not understood about the impacts of CMF on high cloud properties, precipitation rates, and the associated microphysical-dynamical feedbacks. Representation of CMF is also a major source of error in weather and climate models, thereby limiting our ability to predict the microphysical and dynamical properties of convective storms on weather through climate timescales.
INCUS seeks to: (1) identify environmental factors controlling CMF in tropical storms; (2) explore the connection between CMF and high anvil clouds; (3) link CMF to storm type and intensity; and (4) evaluate these relationships in models. The INCUS observations will enhance our understanding and prediction of convective storm processes.
The INCUS mission is the first to systematically measure rapidly changing CMF in tropical convection. It features three SmallSats in low Earth orbit, spaced 30 and 90 seconds apart, each with a Ka-band scanning radar (RainCube heritage). The middle satellite also carries a TEMPEST-D–based passive microwave radiometer. This setup captures radar observations at 30-, 90-, and 120-second intervals, enabling the use of time-differenced radar profiles to retrieve CMF. These observations will help quantify CMF intensity, vertical transport duration, and storm evolution. The radiometer provides insights into high anvil cloud properties and storm context. Together, these instruments will provide unprecedented 3D views of tropical convection.
Extensive research is being conducted in support of the INCUS mission. This includes high-resolution storm simulations, storm tracking, forward modeling, adaptive ground radar scanning, and analysis of storm environments and anvil clouds. This talk will provide an overview of the INCUS mission architecture, time-differencing retrieval approach, and early research results supporting the INCUS science goals related to this session on atmospheric convection.
How to cite: Rasmussen, K., van den Heever, S., Posselt, D., Tanelli, S., Kollias, P., Partain, P., and Stephens, G. and the INCUS Science Team: The INCUS Mission: Measuring Convective Mass Flux from Space, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15871, https://doi.org/10.5194/egusphere-egu26-15871, 2026.
Convective cloud updrafts flux mass vertically throughout the atmosphere and have significant impacts on many atmospheric phenomena, including precipitation production, tropospheric and stratospheric composition, and regional and global circulations. As such, it is important to understand the processes that govern convective cloud updrafts and the scales at which they operate. As part of the NASA Investigation of Convective Updrafts (INCUS) mission, over 60 large-domain, high-resolution simulations have been conducted for convective cloud cases using the Regional Atmospheric Modeling System (2-moment RAMS microphysics) and Weather Research and Forecasting model (2-moment Morrison and Thompson microphysics) in support of the development of the INCUS algorithm and scientific approach. The cases span a wide range of convective storm morphologies from isolated deep convective clouds to mesoscale convective systems and tropical cyclones. The simulations utilize three one-way nested domains with horizontal grid spacings of 1.6 km, 400 m, and 100 m, respectively.
Using the INCUS simulation database, we address the following questions: how much do updraft magnitudes vary as a function of horizontal grid spacing and why? To address these questions, we quantify the differences in vertical velocity between our simulations with 1.6 km, 400 m, and 100 m grid spacing. Initial results show that vertical velocities tend to be stronger below ~ 5 km AGL and weaker above ~5 km AGL, as grid spacing decreases, and that these results are consistent for all three modeling frameworks. We further decompose updrafts into the components of the vertical velocity tendency equation to understand the processes driving vertical velocity differences as a function of grid spacing. With the breadth of the INCUS simulation database, we further quantify how the vertical velocity grid spacing dependency varies as a function of convective system type (e.g. updrafts within scattered convective clouds versus more organized convective systems). This research provides important insights into systematic biases in the representation of deep convective clouds that arise from model horizontal grid spacing and implications for global kilometer-scale modeling.
How to cite: Marinescu, P., Leung, G., Singh, I., Bukowski, J., Grant, L., Storer, R., Rasmussen, K., and van den Heever, S.: Sensitivity of Deep Convective Updraft Magnitudes to Horizontal Grid Spacing: From Kilometer Scales to LES Scales, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15963, https://doi.org/10.5194/egusphere-egu26-15963, 2026.
Marine boundary-layer clouds have been observed to evolve periodically; the cloud field can go through a phase where large, precipitating clouds dominate, followed by a phase where the cloud formation is suppressed by evaporative cooling. We examine the organization and evolution of the marine boundary-layer cloud field, modelled by a high-resolution, large-eddy simulation of a turbulent atmosphere.
Individual cloud regions are isolated and probability distributions of individual cloud properties are used to examine the efficacy of cloud size distribution to represent the strength of convective activities across the observed cloud field. Probabilistic statistical methods based on Bayesian inference are employed to study how the time-series of cloud field properties, such as the cloud size distribution, are correlated to a number of cloud properties, such as individual cloud mass flux. We show that these properties, especially individual cloud mass flux and precipitation, are strongly correlated with the oscillatory changes in the cloud size distribution. The correlations between the distribution of cloud sizes and the strength of turbulent mixing, expressed in terms of direct entrainment and dilution rates, are also examined. These findings have implications for a better representation of convective dynamics in numerical modelling of the atmosphere and a better interpretation of observational data from satellite and in-situ measurements.
How to cite: Oh, G. L. and Austin, P. H.: Representing the oscillatory changes in cloud field properties as a function of cloud size distribution, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16107, https://doi.org/10.5194/egusphere-egu26-16107, 2026.
High-performance computing now enables a new generation of global kilometer-scale models. As part of the World Climate Research Programme (WCRP) Global Hackathon 2025 initiative, for the first time, multiple cutting-edge global kilometer-scale models have been run for an entire year. All of them have covered the summer of 2020, when East Asia experienced record-breaking precipitation and catastrophic floods, mainly driven by mesoscale convective systems (MCSs). Using an updated storm-tracking algorithm, this study investigated the performance of six global kilometer-scale models in simulating MCS characteristics during the record-breaking wet summer of 2020 in East Asia. Results revealed that all models generally reproduced MCS characteristics, including MCS size, duration, and key features of convection and precipitation. Models also generally captured finer characteristics such as diurnal variations and the frequency-intensity distribution of hourly precipitation. Among the models, Integrated Forecast System (IFS) performs best in capturing MCS rainfall spatial distribution, Nonhydrostatic ICosahedral Atmospheric Model (NICAM) excels in simulating MCS size, and Simple Cloud-Resolving E3SM Atmosphere Model (SCREAM) most accurately represents the land-sea contrast in MCS precipitation intensity. A common bias across models is the underestimation of rainfall area and overestimation of heavy precipitation intensity, indicating simulated convective cores are stronger than observed. Our results demonstrate that global kilometer‑scale modeling has reached a significant benchmark, yet persistent biases remain in MCS simulation. Continued improvements in these models will not only enhance the reliability of modeling but also to improve disaster risk reduction and climate change adaptation.
How to cite: Li, P., Chen, H., Li, J., Prein, A., and Zhang, Y.: How Well Do Current Global KM-Scale Models Simulate Storms in East Asia’s 2020 Record-breaking Wet Summer, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16301, https://doi.org/10.5194/egusphere-egu26-16301, 2026.
Trade-wind shallow convection, a major source of uncertainty in cloud-feedback estimates, exhibits pronounced meso-scale (O(10-100 km)) variability, with organization patterns that depart strongly from the canonical “popcorn” cumuli and has thus gained the attention of many investigations in the past decade. We review these advances and present them in a unifying framework which highlights how mesoscale heterogeneities emerge from interactions between internal processes and external forcings. The internal processes include meso-scale circulations, cold-pools and convection-scale elements such as thermals, whereas external forcings are conventional large-scale cloud controlling factors (CCFs) such as sea-surface temperature gradients, wind speed and shear, the stability and humidity at inversion levels, etc. The interactions between these internal processes and external forcings result in a ubiquitous presence of mesoscale organized systems of shallow cumuli which crucially regulate cloud fraction, radiative effects and precipitation. The process-studies over the years allows provides diagnostic means to validate the representation of these processes in models. From literature, we thus also collate diagnostics to be used as a benchmark dataset on meso-scale processes and their large-scale imprint as reference for the validation of simulations, development of new parameterizations and establishment of new constrains. The benchmark dataset is based on high-resolution simulations and observations and contains products and software code to facilitate the inter-comparison with additional sources.
How to cite: George, G., Schulz, H., Alinaghi, P., Chen, X., Denby, L., Eastman, R., Fons, E., Garg, P., Janssens, M., Kruse, I., McCoy, I., Meier, R., Robbins Blanch, N., and Vogel, R.: Mesoscale Processes of Trade-Cumulus Clouds - A Review Framework and Diagnostics for Model Benchmarking, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16448, https://doi.org/10.5194/egusphere-egu26-16448, 2026.
Extreme precipitations from convective storms produce significant harm to society and are expected to become more intense under global warming. Here we examine which storms are conducive to extreme precipitations from the perspective of the mechanical energy expended to build up such storms. Indeed, studies have indicated that about 70% of the mechanical energy of convection is used to lift water, leaving only 30% available for kinetic energy production and the maintenance of convective motions. Here we show that this partitioning holds at the scale of individual storm systems, meaning that the weight of water must have a dramatic impact on storm energy. In particular, storms must build up significant mass by performing significant work against gravity before they can produce the most extreme precipitations. The situation is complicated by the strong diversity across storm systems, characterized by varying durations and varying stages within the lifecycle of those storms. We use kilometer-scale simulations in radiative-convective equilibrium and a convective tracking algorithm to study the mechanical energy budget of storms associated to extreme percentiles of the precipitation distribution. We find that a dominant driver of the storm's ability to produce extreme precipitation is the time span during which the work done to lift water exceeds the dissipation of potential energy through precipitation, thus leading to mass build up inside the storm. Systems generating exceedances of the 99.999th precipitation percentile can accumulate mass during 4-6 hours while more typical precipitating systems only do so over 1-2 hours. Irrespective of storm duration, we find that the fraction of mechanical energy dissipated against gravity stays roughly constant at ~70% throughout the storm lifecycle, creating a drag that must be overcome to produce precipitation extremes. These results improve prospects for more accurate predictions of precipitation extremes from observations.
How to cite: Bolot, M., Fildier, B., Roca, R., Pauluis, O., Fiolleau, T., and Muller, C.: A mechanical energy perspective on the lifecycle of heavy rain-producing storm systems, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19821, https://doi.org/10.5194/egusphere-egu26-19821, 2026.
Convective systems exhibit a wide range of cloud and precipitation structures spanning spatial scales from a few kilometres to thousands of kilometres. While the organization of convection at the meso-alpha scale (200–2000 km) is relatively well-researched through observations and numerical modelling, much less is known about how convection organizes at smaller scales, down to a few kilometres, that are now accessible to kilometre-scale, storm-resolving models.
To address this, we investigate the spatial organization of extreme precipitation in simulations of the storm-resolving model, ICON, coupled to the prognostic aerosol module, HAM-lite. Using month- long, kilometre-scale limited-area simulations over the Atlantic Intertropical Convergence Zone, conducted for the ORCESTRA measurement campaign period [1], we find that 99th-percentile precipitation extremes over the ocean exhibit robust scale-invariant organization across spatial scales from approximately 10 to 150 km, characterised by a fractal dimension of approximately 4/3.
While individual convective updrafts are associated with strong surface convergence, their organisation at these scales is significantly influenced by cold pools which generate intense surface wind divergence. Consistent with this mechanism, grid points with large absolute values of surface wind divergence form spatial clusters that statistically resemble those of extreme precipitation. They tend to predominantly affect the intermittency of surface wind fluctuations, in a manner analogous to shocks in compressible turbulence. Building upon this analogy, we demonstrate that the surface wind fluctuations indeed exhibit a nearly-bifractal scaling — consistent with certain models of compressible turbulence [2] — and the scaling exponents of higher-order surface wind velocity structure functions appear to approach the co-dimension of the fractal set defined by the extreme precipitation events.
This establishes a direct quantitative link between the spatial organization of precipitation extremes and surface wind fluctuations at sub–meso-alpha scales, highlighting implications for the development of simple yet physically grounded stochastic parameterizations of the latter in coarse- resolution GCMs. Furthermore, we assess the robustness of such organization to various climate- change and air pollution scenarios via perturbations to the prescribed sea-surface temperatures and aerosol emissions, respectively.
References:
[1] https://orcestra-campaign.org/intro.html
[2] Mitra et al, Physical Review Letters 94, 194501 (2005).
How to cite: De, S. and Stier, P.: Linking the organization of precipitation extremes at sub-meso-alpha scales to surface wind fluctuations in a storm-resolving GCM , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21026, https://doi.org/10.5194/egusphere-egu26-21026, 2026.
Developing a tractable understanding of the interaction between cumulus cloud tilt and vertical shear, both due to synoptic background winds and neighboring cumulus clouds in a cloud field, is crucial to expanding the theory associated with squall line development and tradewind cumuli climatological feedbacks. Although these interactions are multifaceted, we focus on the dynamic interplay between vertical shear and the cloud-scale flow. To perform this investigation, we implement the Kinematic Representation of Neutrally-buoyant Updraft Tori (KRoNUT) model to represent cloud-scale motions. Unlike previous formulations of the KRoNUT model, we introduce a new tilting parameters into the functional form of the flow. Using a moment closure technique, we then solve for the Dynamics of Neutrally-buoyant Updraft Tori (DoNUT) equations, a coupled non-linear system of ordinary differential equations which govern the temporal evolution of the parameters describing the intensity and geometry of a cloud-scale flow. Using this technique, we analytically and numerically compare the DoNUT equations with and without tilt to determine how the tilting associated with various forms of vertical shear influences the life cycle of a cumulus cloud. When considering the processes of turbulent diffusion and self-advection, we find that tilting alters the nature of a cloud’s steady state circulation. In turn, clouds have the potential to evolve to larger horizontal extents. However, we also find that tilting contributes strongly to enervation and thus the weakening of a cloud’s maximum vertical velocity and a shortening of its life cycle. These impacts are maximized at tilting angles of plus or minus 45 degrees from the vertical axis.
How to cite: Falcone, D., Igel, M., and Biello, J.: A Model for Cumulus Cloud Tilt and the Effect of Vertical Shear, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-22123, https://doi.org/10.5194/egusphere-egu26-22123, 2026.
Posters virtual: Mon, 4 May, 14:00–18:00 | vPoster spot 5
EGU26-19311 | ECS | Posters virtual | VPS2
A Phase-Plane Representation of the Convective Life Cycle: Characterizing Diabatic and Adiabatic Drivers during the Indian Summer MonsoonMon, 04 May, 14:12–14:15 (CEST) vPoster spot 5
Tropical deep convection evolves as a cyclic process, but most observational and modeling studies diagnose convection through regional or domain-based contrasts, obscuring how key physical processes vary across different stages of the convective life cycle. The convective life cycle in the tropics is frequently conceptualized through recharge discharge processes. While valuable, this framework can be extended with more granular, phase specific diagnostics to better understand the distinct physical processes governing each stage of convection. Here, we build on existing phase-plane approaches to represent convection as a cyclic process, using column-integrated moist static energy (MSE) and its temporal tendency as the primary state variables. The phase plane is constructed with column integrated MSE along the horizontal axis and its temporal derivative along the vertical axis. While adhering to the established recharge–discharge paradigm, we extend this terminology by defining four distinct, cycle-consistent stages on the phase plane: Build-up, Cresting, Decay, and Recovery. Applying this framework to the Indian Summer Monsoon (ISM) core region, we map quasi-geostrophic (QG) omega-scaled precipitation components onto the MSE phase plane to investigate the relative contributions of diabatic heating and adiabatic forcing across the convective life cycle. These stage dependent signatures demonstrate the utility of the MSE phase plane for attributing and relative importance of dynamical and diabatic processes across the convective life cycle. Final results and extended analyses will be presented and discussed at the conference.
How to cite: Chakraborty, S., Nikumbh, A., Maithel, V., and Zore, T.: A Phase-Plane Representation of the Convective Life Cycle: Characterizing Diabatic and Adiabatic Drivers during the Indian Summer Monsoon, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19311, https://doi.org/10.5194/egusphere-egu26-19311, 2026.
EGU26-676 | ECS | Posters virtual | VPS2
Radar Polarimetry to Characterize Overshooting Convection in the Western Ghats of IndiaMon, 04 May, 14:15–14:18 (CEST) vPoster spot 5
This study provides the first high-resolution polarimetric radar observations of Overshooting Convective Storms (OCS) over the Western Ghats (WG), India using the newly installed SSPA-based X-band Radar at HACPL, Mahabaleshwar. Three post-monsoon OCS events (15, 23, and 24 October 2025) were analysed using PPI, RHI, CFAD products and ERA5 Atmospheric fields. All storms exhibited strong vertical growth, with echo-top heights of 17.6–19.8 km (15 Oct), 16.5 km (23 Oct), and 17.8 km (24 Oct), and peak reflectivity values of 59.6, 63.3, and 52.1 dBZ, respectively. Notably, significant reflectivity (>40 dBZ) persisted above 16 km, confirming deep overshooting intrusions. Polarimetric signatures showed clear mixed-phase and ice-growth processes, including KDP up to 3–4° km⁻¹, enhanced ZDR in the rainy regions, and reduced ρhv (0.92–0.96) within convective cores, indicating liquid water content, riming, and graupel/hail production. ERA5 diagnostics revealed favorable conditions for deep convection, with strong mid-tropospheric ascent (–0.6 to –0.8 Pa s⁻¹), high moisture, and pronounced convergence over the WG. These results demonstrate intense post-monsoon overshooting convection in complex terrain and highlight the capability of X-band Polarimetric radar to reveal the storm microphysics and vertical structure in an orographically challenging environment.
How to cite: Devisetty, H., Uriya Veerendra, M. K., Tyagi, B., Das, S. K., Chakravarty, K., Survase, C., and Burrala, P. K.: Radar Polarimetry to Characterize Overshooting Convection in the Western Ghats of India, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-676, https://doi.org/10.5194/egusphere-egu26-676, 2026.
