PICO
In this context, this session seeks contributions on all aspects of complexity, nonlinearity, tipping points and stochastic dynamics of the Earth system, including the atmosphere, the hydrosphere, the cryosphere, the solid earth, etc. Communications on theoretical, experimental and modeling studies are all welcome, where the latter modeling studies can span the range of model hierarchy from idealized models to complex Earth System Models (ESM). Studies based on emerging approaches such as data driven models, Artificial Intelligence approaches, complex network methods, critical slowing down analysis, dynamical and stochastic systems theory, etc., are particularly encouraged.
Cascading extreme weather events, characterized by sequential occurrences of distinct extremes such as heatwaves, floods or droughts, pose increasing risks in a warming climate. However, existing approaches for identifying such events focus either on temporal persistence or spatial coherence alone, and are thus unable to identify the most severe events with both characteristics. Here, we propose a new approach based in dynamical systems theory that treats variables as coupled systems, with a view to enable their mechanistic understanding. We illustrate the application of the method to temperature and relative humidity data during the period 1979-2020, identifying cascading heat-drought extremes over the Mississippi, southeastern China and France. While these events are controlled by different large-scale climate modes and blocking patterns, nine of the events occurred during rapid transitions (<12 months) from El Niño to La Niña. In China, these transitional events were consistently preceded by heavy rainfall approximately two weeks earlier. Key drivers include the prolonged presence of the western north Pacific subtropical high and land-atmosphere feedbacks. Our findings uncover the speed and severity of cascading wet-dry transitions within as little as two weeks during El Niño transition years, and the need for a greater understanding of their driving mechanisms.
How to cite: Deng, Q., Slater, L., Franzke, C., Guo, Y., and Fu, Z.: Whiplash weather in ENSO Transition Years Identified by A Novel Cascading Extremes Index, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8761, https://doi.org/10.5194/egusphere-egu26-8761, 2026.
The Atlantic Multidecadal Variability (AMV) and the North Atlantic Oscillation (NAO) are the dominant modes of oceanic and atmospheric variability in the North Atlantic, respectively, and are key sources of predictability from seasonal to decadal timescales. However, the physical processes and feedback mechanisms linking the AMV and NAO, and the role of diabatic processes in these feedbacks, remain debated. We present a data-driven dynamical modelling framework which captures coupled decadal variability in AMV, NAO, and North Atlantic precipitation. Applying equation discovery methods to observational data, we identify deterministic low-order dynamical models consisting of three coupled ordinary differential equations. These models reproduce observed North Atlantic decadal variability and show robust out-of-sample predictive skill on multi-annual to decadal lead times. The resulting model dynamics include a distinct quasi-periodic 20-year oscillation consistent with a damped oceanic mode of variability. Notably, precipitation-related terms feature prominently in the low-order models, suggesting an important role for latent heat release and freshwater fluxes in mediating ocean–atmosphere interactions. We propose new feedback mechanisms between North Atlantic sea surface temperature and the NAO, with precipitation acting as a dynamical bridge. By linearising the low-order models and computing finite-time Lyapunov exponents, we find that North Atlantic precipitation is more predictable in a positive AMV phase. We then analyse several decadal prediction ensemble experiments based on initialised hindcasts and find comparable state-dependent predictability of precipitation. Overall, these results illustrate how data-driven equation discovery can provide mechanistic hypotheses and new insight beyond conventional analyses of observations and climate model simulations.
How to cite: Nicoll, A., Christensen, H., Huntingford, C., and Smith, D.: New insights into decadal climate variability in the North Atlantic revealed by data-driven dynamical models, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3111, https://doi.org/10.5194/egusphere-egu26-3111, 2026.
Ensemble forecast generate multiple predictions from a set of initial conditions, thereby producing the probability density distribution (PDF) of a variable and quantifying forecast uncertainty beyond a single deterministic forecast. However, studies focusing on the predictable lead time of ensemble forecast remain limited. In this study, orthogonal conditional nonlinear optimal perturbations (O-CNOPs) are applied to the Lorenz-96 model to investigate the predictable lead time of ensemble forecast, which is then compared with that obtained from a single deterministic forecast. Results show that the maximum predictable lead time revealed by the ensemble distribution generated with O-CNOPs is 18.5 days, 2.5 days longer than that revealed by the ensemble distribution generated with singular vectors (SVs), which is 16 days. Consistent results are obtained from the ensemble mean analysis, which reveals a longer predictable lead time for O-CNOPs (21.75 days) than for SVs (18 days). In addition, compared with ensemble forecasts generated with SVs, the ensemble forecasts generated with O-CNOPs exhibit higher deterministic forecast skill, probabilistic forecast skill, reliability, and resolution over the same forecast periods. These results collectively highlight the advantage of O-CNOPs in constructing physically consistent nonlinear ensemble distributions and provide a methodological framework for more accurate quantification of ensemble forecast lead time.
How to cite: Zhu, Y. and Duan, W.: Exploring the Predictable Lead Time of Ensemble Forecast Based on Conditional Nonlinear Optimal Perturbation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3320, https://doi.org/10.5194/egusphere-egu26-3320, 2026.
Climate oscillations such as the El Niño–Southern Oscillation (ENSO) and the Madden–Julien Oscillation (MJO) dominate aspects of climate variability, yet they are often challenging to accurately capture in climate models. Due to their disparate underlying physical processes, any potential commonality between different climate oscillations is obscured. Common underlying dynamics is suggested by the success of relatively low-dimensional linear inverse modeling (LIM). LIMs represent climate oscillations as linear Gaussian nonequilibrium steady states (LG-NESS) defined by stochastic differential equations. Here we develop the theory of LG-NESS’s and compare with observations and models of climate oscillations.
ENSO and the MJO are often described by two-dimensional indices such as the leading SST EOFs for ENSO, or the Realtime Multivariate MJO index. The LIM algorithm parameterizes the dynamics in the index coordinate system as a two-dimensional LG-NESS specified by seven parameters. We decompose the parameter space into four parameters that define the coordinate system of the index, and three parameters that define its intrinsic dynamics. This allows us to transform all 2d LG-NESS’s to a common three-dimensional dynamical parameter space. Coordinate-invariant quantities depend only on the three dynamical parameters, while coordinate-dependent quantities can be transformed back to the original index coordinate system and depend on all seven parameters.
We parameterize ENSO and the MJO in this three-dimensional dynamical parameter space and find that, despite their distinct physical mechanisms and timescales, they lie within a narrow region of parameter space, indicating a similarity in the underlying phase-space dynamics. We compare observed and modeled dynamics with those of their parameterized LG-NESS, evaluating predictability, thermodynamic properties, and event statistics. We find this minimal three-parameter model reproduces many features of climate oscillations, revealing a deep dynamical similarity among climate oscillations.
How to cite: Weiss, J., Benincasa, R., Du, D., and Duane, G.: Climate Oscillations and Linear Gaussian Nonequilibrium Steady-States, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8143, https://doi.org/10.5194/egusphere-egu26-8143, 2026.
The skill of forecasting Tropical Cyclone (TC) Rapid Intensification (RI) is limited by inherent uncertainties in initial conditions and model physics. To address this, the C-NFSVs method integrates initial and model perturbations, accounting for their collective effects through the nonlinear forcing singular vector (NFSV; also known as CNOP-F) approach. In this study, we applied C-NFSVs to the Weather Research and Forecasting (WRF) model for TC ensemble forecasting across three resolutions, comparing it against O-NFSVs, which has proven superior to traditional stochastic physics schemes. Results reveal a significant resolution dependence, with the superiority of C-NFSVs maximizing at the convection-permitting scale. At this resolution, the C-NFSVs ensemble outperforms O-NFSVs for both deterministic and probabilistic metrics, and demonstrates significantly improved reliability. Notably, for the challenging prediction of RI events, C-NFSVs exhibits high discriminative skill, achieving an Area Under the ROC Curve (ROCA) of 0.80. A detailed examination of TC Hato attributes this success to capturing the evolution of the critical physical error chain, which progresses from thermodynamic priming and convective organization to the structural and dynamic response. Mechanistically, the results highlight the complementary roles of the two components: the initial component of C-NFSVs dominates the uncertainty of the dynamic structure in the early forecast stage, while the model component plays a primary role in maintaining the thermodynamic uncertainty of moisture and temperature fields throughout the forecast. This study validates the effectiveness and physical rationality of C-NFSVs in high-resolution ensembles, offering a promising strategy for enhancing the predictability of extreme weather events at convection-permitting scales.
How to cite: You, C. and Duan, W.: Enhancing Tropical Cyclone Ensemble Forecast Skill via the Collective Effect of Initial and Model Perturbations: The C-NFSVs Method, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3326, https://doi.org/10.5194/egusphere-egu26-3326, 2026.
Network-based early warning signals of El Niño have been recognized for more than a decade, however, it remains unclear whether current climate models can reproduce these signals. Here, we evaluate simulations from both the pre-industrial control and historical experiments of CMIP6 models. While none of the models exhibited skill in either experiment, performance was generally better in the historical runs, suggesting that the inclusion of external forcing may improve model simulations of the early warning signals. Further analysis indicates that some models such as CESM2, FGOALS-g3, and MRI-ESM2-0 may provide potentially useful early warning information for El Niño events, but their warning signals tended to emerge later than those in reanalysis data. Using a new network-based evaluation metric to assess air-sea interactions in the tropical Pacific, we find that model performance in simulating early warning signals is generally linked to their ability to simulate these interactions. This highlights the importance of improving representations of air-sea coupling in current models. For future investigations into the physical mechanisms underlying the network-based early warning signals, CESM2, FGOALS-g3, and MRI-ESM2-0 are recommended due to their relatively better performance compared to the other models considered in this work, although the causes of their delayed signal emergence require further exploration.
How to cite: Yuan, N., Han, J., and Ludescher, J.: Evaluation of CMIP6 Models in Simulating Network-Based Early Warning Signals of El Niño, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10698, https://doi.org/10.5194/egusphere-egu26-10698, 2026.
The growing availability of multiple operational ocean data services provides unprecedented opportunities for applications such as environmental incident response, search and rescue operations, and maritime management. At the same time, despite their widespread use, most ocean datasets offer limited information regarding their performance and consistency with real-world observations.
In this presentation, I address this gap by introducing a methodology to assess uncertainty in ocean transport predictions derived from different ocean data products. Building on recent work that links transport uncertainty—understood here as deviations from ground truth—to invariant dynamical structures in the ocean [1–3], the proposed approach, discussed in [4], exploits these links to guide statistical averaging strategies. We examine how well model-predicted material transport aligns with observational evidence across different dynamical scales, including scales above the mesoscale, the mesoscale, and the submesoscale. This perspective provides a systematic pathway for quantifying the performance of different data sources and assessing their overall quality and reliability.
References:
[1] G. García-Sánchez, A. M. Mancho, A. G. Ramos, J. Coca, B. Pérez-Gómez, E. Alvarez-Fanjul, M. G. Sotillo, M. García-León, V. J. García-Garrido, S. Wiggins. Very High Resolution Tools for the Monitoring and Assessment of Environmental Hazards in Coastal Areas. Frontiers in Marine 7, 605804 (2021).
[2] G. García-Sánchez, A. M. Mancho, S. Wiggins. A bridge between invariant dynamical structures and uncertainty quantification. Commun. Nonlinear Sci. Numer. Simul. 104, 106016 (2022).
[3] G. García-Sánchez, A. M. Mancho, M. Agaoglou, S. Wiggins. New links between invariant dynamical structures and uncertainty quantification. Physica D 453 133826 (2023).
[4] G. García-Sánchez, M. Agaoglou, E.M.C Smith, A. M. Mancho. A Lagrangian uncertainty quantification approach to validate ocean model datasets. Physica D 475 134690 (2025).
Acknowledgments:
Support from PIE project Ref. 202250E001 funded by CSIC, from grant PID2021-123348OB-I00 funded by MCIN/ AEI /10.13039/501100011033/ and by FEDER A way for making Europe.
How to cite: Mancho, A. M.: Understanding Uncertainty in Ocean Transport Inferred from Multiple Data Sources, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13663, https://doi.org/10.5194/egusphere-egu26-13663, 2026.
The magnetic field of planets and stars is generated by the movement of conductive fluids inside these bodies. The precession and libration of these astrophysical bodies play a central role in the excitation of the internal turbulent fluid motion. In our laboratory, we have developed an experiment that allows the investigation of precession-driven inertial waves and their instability (Xu and Harlander, 2020). Wave triads play a very important role in this instability (Lagrange et al., 2011). As the Ekman number decreases, an increasing number of interacting triads arise, ultimately leading to turbulence. This process can be experimentally reproduced in the laboratory. In this experiment, precession is simulated using a slightly tilted cavity with a free fluid surface and is therefore simpler in design than a real precession experiment.
The dynamics of fluids can be described by PDEs. However, often deeper insights can be gained from a corresponding low-dimensional dynamical system. An example is the large family of Lorenz-type models, which have led to a fundamental understanding of predictability in atmospheric dynamics (Majda et al., 1999). Also, for the problem of a precessing rotating cylinder, low-dimensional models exist. Such models are obtained from spectral discretizations of the Navier-Stokes equations and truncating the resulting hierarchy of coupled equations at low order. Truncation, however, eliminates the quadratic coupling between the resolved modes and the (unresolved) smaller scales, which can lead to unrealistic characteristics of turbulence.
We suggest another closure to systematically derive low-order amplitude equations for rotating fluids, based on stochastic modeling of the unresolved small scales in accordance with the available experimental data. Specifically, we first remodel the small scales by an appropriate stochastic process that has a multivariate Gaussian law when conditioned on the resolved variables and, in a second step, apply a projection operator to the coupled system. In doing so, we derive closed, averaged equations for the resolved variables that retain the quadratic nonlinearities and so capture the small-scale contributions to the low-order wave dynamics. For a projection operator in the form of a conditional expectation (i.e., a projection on function space), we have recently studied necessary and sufficient conditions under which the projection operator formalism yields an approximation for nonreversible (e.g. driven) systems (Duong et al., 2025). Measuring the distance between the marginal distributions of the resolved variables for the full- and the low-order models, the accuracy of the low-order model can be measured (Hartmann et al., 2020).
By comparing the low-order stochastic model results with data from the precession experiment, the hope is not only to capture the wave interactions correctly and develop a stochastic extension of the existing amplitude equations, but also to reduce the order of the existing model even further.
M.H Duong, C. Hartmann, and M. Ottobre, arXiv preprint, arXiv:2506.14939, 2025.
C. Hartmann, L. Neureither, and U. Sharma, SIAM J. Math. Anal. 52(3), 2689-2733, 2020.
R. Lagrange, P. Meunier, F. Nadal, C Eloy, J. Fluids Mech. 666, 104–145, 2011.
A.J. Majda, I. Tomofeyev, E. Vanden Eijnden, PNAS, 96(26), 14687-14691, 1999.
W. Xu, U. Harlander, Rev. Phys. Fluids., 5(9), 094801-21, 2020.
How to cite: Harlander, U. and Hartmann, C.: Low-dimensional stochastic amplitude equations for a precessing rotating cylinder, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5335, https://doi.org/10.5194/egusphere-egu26-5335, 2026.
In the Model Uncertainty-Model Intercomparison Project (MUMIP) we compare parameterisation packages from different modelling centres using their single-column modelling (SCM) frameworks. We will showcase the dataset from an Indian Ocean experiment at a 0.2 degrees grid covering one month, with about 10 million simulations of each model. These parametrised models are compared against a convection-permitting benchmark from DYAMOND under common dynamical constraints. We will show differences and similarities in precipitation patterns and physics tendencies among four models and show how these differences can be generalised. Following earlier works, we find that at coarse grids that do not resolve convection, parameterisation packages tend to produce overconfident tendencies compared to the convection-permitting benchmark. Furthermore, we test several hypotheses on the MUMIP dataset to explain the differences. We use the data to explore the foundations of stochastic physical parametrisations. Would stochastic physics effectively overcome the overconfidence for good reasons? May the stochastic perturbations actually have a physically meaningful quantitative interpretation? Can stochastic physics be used to partially overcome truncation and grid spacing limitations?
How to cite: Groot, E., Christensen, H., Sun, X., Newman, K., Lfarh, W., Roehrig, R., Bengtsson, L., and Simonson, J.: How different are parameterisation packages really and how can we interpret stochastic perturbations?, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2991, https://doi.org/10.5194/egusphere-egu26-2991, 2026.
This study investigates how complex volcanic systems undergo major behavioral shifts, focusing on the Solfatara–Pisciarelli (SP) hydrothermal-magmatic area within the Campi Flegrei caldera (Southern Italy). The SP system is one of the most active zones of the caldera, characterized by persistent degassing, seismic swarms, strong hydrothermal circulation and long-term ground uplift. These processes arise from nonlinear interactions between magmatic inputs, fluid migration, and shallow hydrothermal pressurization, making the identification of critical transitions particularly challenging.
To address this, we developed an integrated analytical framework combining Multivariable Fractional Polynomial Analysis (MFPA) and Global Critical Point Analysis (GCPA). MFPA models nonlinear and time-lagged associations among key monitoring parameters—vertical ground deformation, seismicity, CO₂ flux, geochemical equilibrium variables, and thermal signals—while GCPA identifies the temporal moments when multiple variables collectively show systemic reorganization.
Analysis of multi-year (2018–2024) geophysical and geochemical datasets revealed that deformation is strongly associated with seismicity, equilibrium pressures of hydrothermal gases, heat flow, and CO₂ flux. Incorporating time-lagged deformation improved model accuracy and reduced unexplained variance, highlighting delayed cause–effect couplings between deformation and fluid-dynamic processes. The model confirms seismicity as the most stable explanatory parameter, consistent with sustained fracturing and fluid pressurization in the shallow system.
GCPA identified two major critical transitions:
- CP1 – 30 November 2020, dominated by thermal–chemical reorganization and increased gas-system pressurization.
- CP2 – 1 April 2023, reflecting a more open and multiparametric regime where deformation, temperature, seismicity, heat flux, and CO₂ emissions contribute comparably to system evolution.
These transitions align with independent geodetic evidence suggesting migration and reconfiguration of the shallow overpressure source beneath the SP area. The integrated MFPA–GCPA approach thus reconstructs how systemic changes propagate across geophysical and geochemical variables, providing retrospective insight into the onset and progression of unrest phases.
This framework offers several advantages over classical or non-parametric approaches: interpretability of functional relationships, explicit treatment of nonlinearities and time lags, and the ability to detect collective regime shifts rather than isolated anomalies. Although not predictive, the method provides a quantitative basis for identifying critical phases in volcanic systems and may be adapted to other densely monitored calderas. With higher-resolution and real-time data streams, it could support early indications of evolving unrest and enrich next-generation volcano-monitoring strategies.
How to cite: Vitale, A., Barone, A., Marotta, E., Vitale, D. F., Pepe, S., Peluso, R., Castaldo, R., Avino, R., Mercogliano, F., Pepe, A., Accomando, F., Avvisati, G., Belviso, P., Bellucci Sessa, E., Antonio, C., Maddalena, P., Sansivero, F., and Tizzani, P.: Critical Transitions at Campi Flegrei Resurgent Caldera: A Novel Approach to Systemic and Retrospective Signals Analysis, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-673, https://doi.org/10.5194/egusphere-egu26-673, 2026.
Understanding the long-term dynamics of forest aboveground carbon (AGC) is critical for constraining the terrestrial carbon cycle. However, accurately reconstructing historical AGC spatiotemporal patterns remains a challenge due to the complex, nonlinear relationships between vegetation proxies and biomass, as well as the stochastic uncertainties inherent in multi-source satellite observations.
In this study, we propose a probabilistic deep learning framework to reconstruct harmonized, high-resolution (0.25°) global forest AGC stocks and fluxes from 1988 to 2021. By integrating multi-source optical (e.g., NDVI, LAI) and microwave (e.g., VOD) remote sensing data, our approach utilizes Probabilistic Convolutional Neural Networks (CNNs) to simultaneously estimate AGC dynamics and quantify associated predictive uncertainties (decomposing aleatoric and epistemic components). This data-driven model effectively captures the nonlinear spatial dependencies and texture features that traditional empirical methods often miss.
Our reconstruction reveals significant decadal-scale regime shifts in the global forest carbon sink. While global forests remained a net sink of 6.2 PgC over the past three decades, we identify a pronounced transition in moist tropical and boreal forests, which have shifted from carbon sinks to sources since the early 2000s. Furthermore, our analysis uncovers an intensifying negative coupling between interannual tropical AGC fluxes and atmospheric CO2 growth rates (r=-0.63 in the last decade), suggesting a growing complexity in the climate-carbon feedback. Spatially explicit partitioning in the Amazon further indicates a dynamical shift where AGC losses are increasingly driven by indirect climate stressors in previously "untouched" forests, rather than direct deforestation alone.
In conclusion, this study elucidates the state-dependent responses of global forests to changing disturbance regimes. The probabilistic framework provides a necessary basis for distinguishing genuine regime shifts, such as the structural decline of the tropical carbon sink, from observation noise, thereby enhancing our predictive understanding of terrestrial carbon resilience in a warming climate.
How to cite: Qian, Z., Bathiany, S., Liu, T., Blaschke, L., Teo, H. C., and Boers, N.: Reconstruction of Global Forest Aboveground Carbon Dynamics with Probabilistic Deep Learning, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4547, https://doi.org/10.5194/egusphere-egu26-4547, 2026.
While understanding systemic risk in complex systems has gained growing attention, less effort is often dedicated to understanding the system itself. Particularly, the typology of complex systems and collapse mechanisms that are consistent across domains remains understudied. Hence, critical questions arise, such as what do we need to know about the system’s characteristics to predict systemwide collapse, identify leverage points, or design resilience interventions? What system properties allow knowledge gained from one system to be generalized to other taxonomically similar systems? What signals can be deduced from a few systems’ global parameters to determine whether a system is in a stable, unstable, or critical region of its adjacent becoming?"
Answering these questions requires determining the typology of complex systems, which enables the study of system-level behaviors independently of the specific details of individual agents. This leads to universality, facilitating the study of collapse mechanisms transferable to other typologically similar systems, thereby providing insight into systemic risk.
This presentation introduces a novel typology of complex systems based on the concept of “adjacent becoming,” drawing on works of Stuart Kaufmann, C.S. Holling, and Marten Scheffer, among others which have established the language of attractors, regime shifts, evolution, and panarchical resilience in complex systems. The System’s Adjacent Becoming (SAB) is what the system is positioned to become while appearing to be in a stable condition, i.e., potential for a critical transition in deep stability. Such a proximal transformation potential can be characterized by four interrelated components consisting of a) the system's location in phase space and proximity to the most accessible alternative attractors, b) the topography of the current boundary basin, c) the system's current momentum and energy state, and d) the prospective trajectory and regime that a transition to a given alternative attractor would induce. These four components collectively determine the SAB potential, and thus the likelihood and qualitative characteristics of an imminent regime shift.
To assess SAB, what system has, what system does, and what system could become are the critical questions. For such an assessment, a SAB-informed typology would be the first step. Therefore, the four SAB components lead to types based on nine interconnected system variables: (1) micro-macro dynamic type; (2) state of information processing and memory capacity; (3) degree of teleonomic coherence across levels and panarchical organization; (4) degree of agent heterogeneity; (5) type and intensity of emergence; (6) functional and computational efficiency rate; (7) initial condition and presence of path dependency; (8) manifestation of critical slowing down indicators and bifurcation proximity signals and (9) the existing geometric attractor landscape.
This SAB-informed typology is phenomenologic-mechanistic in nature, which helps to learn about the structural and dynamical signatures of critical transitions and the quality of the new becoming, offering a unified language for understanding how complex adaptive systems of any kind approach their adjacent becoming and what determines whether they persist, transform, or collapse. This framework remains theoretical with operationalization challenges; future work must advance toward measurable proxies for the nine categories to quantify SAB of real-world systems.
How to cite: Zamanifar, M. and Samaro, N.: Systemic risk in complex systems: understanding the system based on the system’s adjacent becoming, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21170, https://doi.org/10.5194/egusphere-egu26-21170, 2026.
