EGU General Assembly 2024  

How to cite: Pouquet, A.: On a few characteristics of geophysical turbulent flows , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10769, https://doi.org/10.5194/egusphere-egu24-10769, 2024.

During the last decades, space missions provided in situ data of diverse space plasma environments with increasingly higher resolution. This enabled the possibility to investigate peculiar properties of fluctuations in the magnetic field and plasma parameters, transitioning from the magnetohydrodynamic (MHD) to the ion-kinetic regime. The ion-kinetic regime is characterized by a global self-similar scaling of fluctuations, in contrast to the local scale-invariance of the MHD ones. In a series of works, we developed a data-driven approach based on the Langevin equation in order to model statistical features of kinetic fluctuations. In practical terms, the stochastic process thus introduced represents the evolution of the magnetic field fluctuations as a function of the scale. As far as such fluctuations are of the Langevin type, their statistics evolve according to a Fokker-Planck equation. Studying the evolution of fluctuation statistics across the scales, e.g., structure functions, allows us to make predictions about global statistical properties, e.g., scaling exponents.

In this work, we review recent results obtained by using data from the ESA/Cluster mission in near-Earth space. We give evidence that the dynamics of magnetic field increments at kinetic scales can be modeled as a stochastic process of the Langevin type, and that the correct scaling law of the structure functions can be obtained through the stochastic equation in the non-diffusive limit, by linking the drift term of the Langevin equation to the Hölder exponent. Finally, this model allows us to derive the asymptotic limit of individual sets of fluctuations, giving thus predictions on the trend expected at kinetic scales.

How to cite: Benella, S.: Exploring space plasma fluctuations at kinetic scales through stochastic process theory, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11136, https://doi.org/10.5194/egusphere-egu24-11136, 2024.

Medicanes, tropical-like cyclones in the Mediterranean Sea, pose unexpected challenges to unprepared areas due to their projected increases in intensity. To address these challenges, we proposed: 1) the development of an automatic tracking method in the absence of a comprehensive tracking database for Medicanes; 2) the implementation of a forecasting model for extreme cyclones utilizing artificial intelligence techniques. This is especially beneficial when traditional numerical models struggle to account for nonlinear interactions. We use a K-means algorithm and mean sea level pressure reanalysis data to track storm centers, determining maximum wind speed and position throughout each case’s lifetime. This information categorizes our dataset into storm-like and extreme Medicanes, and facilitates the extraction of spatiotemporal data from infrared satellite images. These features enable us to predict the final classification of Medicanes (whether they are storm-like or extreme) 6 to 36 h before peak wind speed, using an optimized combination of Convolutional Neural Network and Random Forest binary classification methods. By training and testing on Mediterranean data from 1984 to 2020, we successfully diagnosed between 72% and 87% of extreme Medicanes in the studied cases, depending on the lead-time. Our study is the first to employ artificial intelligences for both tracking and forecasting Medicanes, offering a foundational approach to enhance Medicanes preparedness and awareness.

How to cite: Martinez-Amaya, J., Nieves, V., and Muñoz-Marí, J.: Advancements in Medicanes Tracking and Forecasting Using Artificial Intelligence, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-119, https://doi.org/10.5194/egusphere-egu24-119, 2024.

A one-dimensional energy balance model (1D-EBM) is a simplified climate model that describes the evolution of Earth's temperature based on the planet's energy budget.

In this study, we examine a 1D-EBM that incorporates a parameter representing the impact of carbon dioxide on the energy balance. Based on empirical studies showing that bistability may occur in Earth's tropics, we consider the planet's ongoing radiation to be latitude-dependent, presenting bistability in low-latitude regions but not in high-latitude ones. This local bistability does not lead to a bifurcation in the entire system, in addition to the classical saddle-node bifurcations between Snowball Earth and the present climate.

We focus on investigating the statistical properties of the system when the model is perturbed with additive noise. Our work is a step towards a clearer understanding of the dynamics in a spatially heterogeneous setting.

How to cite: Del Sarto, G. and Flandoli, F.: Statistical properties for a spatially heteregeneous one-dimensional energy balance model perturbed by additive noise, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1008, https://doi.org/10.5194/egusphere-egu24-1008, 2024.

The introduction of random perturbations by noise in partial differential equations has proven extremely useful to understand more about long-time behaviour in complex systems like atmosphere and ocean dynamics or global temperature. Considering additional transport by noise in fluid models has been shown to induce convergence to stationary solutions with enhanced dissipation, under specific conditions. On the other hand the presence of simple additive forcing by noise helps to find a stationary distribution (invariant measure) for the system and understand how this distribution changes with respect to changes in model parameters (response theory). I will discuss these approaches with a multi-layer quasi-geostrophic model as example.

How to cite: Carigi, G.: Long-time behaviour of stochastic geophysical fluid dynamics models., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1133, https://doi.org/10.5194/egusphere-egu24-1133, 2024.

In this contribution, we will analyse simple data assimilation schemes that not only estimate the underlying states of a dynamical system but simultaneously reconstruct unknown components of the dynamics. We focus on quasigeostrophic and transport-diffusion equations (for instance for atmospheric aerosols or tracer gases) and reconstruct forcings or surfacd fluxes, along with the underlying dynamical states. Tracer gases and aerosols play an important role in the dynamics of the atmosphere; aerosols for instance act as condensation nuclei and thus have a major influence on precipitation, while tracer gases such as ozone, methane, or CO₂ impact the radiative transfer and are thus linked to important atmospheric phenomena such as the ozone hole and the energy budget of the planet ("greenhouse effect"), respectively. Furthermore, gases as well as aerosols (especially in the lower troposphere) are common pollutants with strong and potentially adverse effects on the environment, human activity, and health. We discuss two algorithms that both apply in the context of the quasigeostrophic as well as the transport-diffusion equations.

How to cite: Bröcker, J., Carigi, G., Kuna, T., and Martinez, V. R.: Data assimilation and the reconstruction of surface fluxes in quasigeostrophic and transport equations., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1148, https://doi.org/10.5194/egusphere-egu24-1148, 2024.

Hydrostatic models were and still are the workhorses for realistic simulations of the ocean dynamics, especially for climate applications. The hydrostatic approximation is formally first order in $\gamma=H/L$, where $H$ is the vertical and $L$ the horizontal scale of the phenomenon considered. For stratified rotating flow the dynamics can be separated in balanced flow and wave motion. It is shown that for the linear balanced motion the hydrostatic approximation is exact and for wave motion it is second order, obtaining the leading prefactors. The validity of the hydrostatic approximation therefore also relies on the ratio of the amplitude of wave motion to balanced motion. This ratio adds considerably to the quality of the hydrostatic approximation for larger scale flows in the atmosphere and the ocean.
  
Imposing the divergenceless condition is a linear projection of the dynamical variables in the subspace of divergenceless vector fields, for both the Navier-Stokes and the hydrostatic formalism. Both projections are local in Fourier space.
  Calculating the difference of the two projections, the expression of the error, scaling and prefactors, done by the hydrostatic approximation is obtained. Analyzing the eigen-space of the projector, it is shown that for rotating-buoyant vortical-flow the hydrostatic-approximation is of third order for buoyant forcing, second order for horizontal and first order for vertical dynamical forcing.
  
Using the Heisenberg-Gabor limit it is shown that for large scale ocean dynamics, the difference of the dynamics of the projection-evolution operator between the two formalisms is insignificant. It is shown that the hydrostatic approximation is appropriate for realistic ocean simulations with vertical viscosities larger than $\approx10^{-2}$m$^2$s$^{-1}$. A special emphasis is on unveiling the physical interpretation of the calculations.
  

How to cite: Wirth, A.: On the hydrostatic approximation in rotating stratified flow, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1469, https://doi.org/10.5194/egusphere-egu24-1469, 2024.

Assimilating dynamic models and observations, along with their errors using Bayesian estimation method are challenged when the model has both aleatoric and epistemic errors. We devised a diffusion map technique that can filter an observational data stream, stripping it of components that are near statistically stationary, leaving behind what we denote the tendency of the time series. The tendency of the time series can be thought of as an executive summary of the time series. A model constructed on known physical principles may not be able to capture the tendency with fidelity and thus one can identify, from an estimation process on aleatoric fields, the epistemic error. Using machine learning strategies a surrogate model for the epistemic error can be inferred from a comparison of the physics model and the tendency. The surrogate model is thus incorporated into the model dynamics to enhance the fidelity in predictions. The assimilation of the enhanced model and the observations can now be carried out over the aleatoric Bayesian framework. To meet the challenge of the resulting highly nonlinear and non-Gaussian data assimilation we employ a newly developed Stein sampler we call the particle flow filter. In this talk we will describe and demonstrate  this assimilation strategy. 

How to cite: Restrepo, J., Ramirez, J., vanLeeuwen, P., and Alves, C.: Data Assimilation with Biases & Random Errors, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2404, https://doi.org/10.5194/egusphere-egu24-2404, 2024.

El Niño-Southern Oscillation (ENSO) is a complex climatic phenomenon with significant impacts on global weather patterns and ecosystems. Improving ENSO predictability is therefore an issue of high societal value. However, Global Circulation Models present severe biases when predicting ENSO, and their skill remains comparable to that of vastly simpler empirical models such as Linear Inverse Models (LIMs). LIMs, however, rely on linear dynamics, and they have inherent limitations in capturing the behavior of non-linear phenomena. In this context, Koopman operator theory has emerged as a powerful mathematical framework, offering a novel perspective for analyzing complex non-linear systems, such as ENSO. In this study, we investigate the potential of Koopman operator theory to enhance ENSO forecasting accuracy. Leveraging 2000 years of tropical SST pre-industrial CESM data, we have assessed the skill of the Niño 3.4 index forecasts using the Koopman framework, and compared it to the benchmark set by LIMs. Our analysis includes sensitivity testing of both methods across various parameters, such as retained variability and data length used for operator computations. Our findings reveal nuances in the robustness of Koopman Operator estimates, particularly evident when using shorter training periods, contrasting with more stable LIM counterparts. However, a notable breakthrough emerges as we demonstrate the higher skill of Koopman multimodel ensembles, showcasing consistent improvements over linear models. The comparative analysis highlights the potential of Koopman operator theory in advancing ENSO forecasting beyond linear models. The utilization of Koopman multimodel ensembles emerges as a promising strategy, demonstrating enhanced forecasting capabilities. Yet, challenges in robustness persist, particularly in shorter data spans, signaling avenues for further refinement. Overall, these findings underscore the significance of the Koopman framework and lay the groundwork for future research aimed at refining methodologies for more accurate predictions in complex climatic systems.

How to cite: Lorenzo Sánchez, P., Newman, M., Navarra, A., Albers, J., and Subramanian, A.: Koopman operator theory for enhanced Pacific SST forecasting, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2554, https://doi.org/10.5194/egusphere-egu24-2554, 2024.

A refined 4D-Var assimilation system within DestinE allows us to assimilate the Meteosat-10/SEVIRI clear-sky radiances over Europe, as well as GOES-16/ABI and GOES-18/ABI, or HIMAWARI/AHI globally at a spatial scale of 75 km instead of the previous 125 km in the ECMWF Integrated Forecasting System (IFS). Higher resolution observations can potentially improve the analysis and therefore the prediction of extreme weather events over Europe, as well as globally. The effects of using higher resolution observations have been investigated with a detailed set of experiments and the impact on wind, temperature, and humidity has been evaluated. A broad range of experiments indicate that exploiting the higher spatial density clear-sky radiances leads to an improvement of humidity sensitive fields in short-range forecasts with the IFS as independently measured for example by instruments on low-earth-orbiting satellites (IASI, CrIS, SSMIS, or ATMS). Due to a reduced displacement and representativeness error, these changes could further lead to improvements in longer range forecasts as these errors propagate upscale nonlinearly. However, so far the impact on the medium range has been mostly neutral.

In addition, pre-processed GOES-16/ABI and GOES-18/ABI observations by NOAA have been assimilated with 10 min sampling rates at 75 km spatial density. Exploring how to best assimilate relatively small spatial and temporal scales for one geostationary satellite, will allow us to approach these smaller scales with other satellites such as HIMAWARI/AHI above the Pacific or MTG-I/FCI above Europe. Data from both satellites will be available for us early in 2024. Preliminary experiments demonstrate the ability of IFS to assimilate observations at the highest available temporal resolution for the GOES-16 and GOES-18 satellites. Higher resolution radiances observed at these shorter time intervals naturally capture smaller scale atmospheric features such as mesoscale convective systems. In our experiments, simultaneously assimilating observations at a higher spatial and temporal resolution leads to an impact that is only marginally better than assimilating higher density observations alone, suggesting a combined investigation of optimal time-assignment, as well as assessment of the observation error are needed to optimise the integration of rapid update measurements in 4D-Var. 

How to cite: Schröttle, J., Lupu, C., and Burrows, C.: Approaching the sub-mesoscale globally at 10 min temporal resolution through assimilating radiances measured by geostationary satellites, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3222, https://doi.org/10.5194/egusphere-egu24-3222, 2024.

How to cite: Toth, Z. and Jankov, I.: Initial-Value vs. Model-Induced Forecast Errors: A New Perspective, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4179, https://doi.org/10.5194/egusphere-egu24-4179, 2024.

With the growing complexity of land surface models used to represent the terrestrial part of wider Earth system models, the need for sophisticated and robust parameter optimisation techniques is paramount. Quantifying parameter uncertainty is essential for both model development and more accurate projections. History matching is an emerging technique in climate science for uncertainty quantification. Using Gaussian process emulators, history matching allows us to rule out parts of parameter space that lead to model outputs being inconsistent with observations. In this presentation, we assess the power of history matching by comparing results to variational data assimilation, commonly used in land surface models for parameter estimation. Although both approaches have different setups and goals, we can extract posterior parameter distributions from both methods and test the model-data fit of ensembles sampled from these distributions. Using a twin experiment, we test whether we can recover known parameter values. Through variational data assimilation, we closely match the observations. However, the known parameter values are not always contained in the posterior parameter distribution, highlighting the equifinality of the parameter space. In contrast, while more conservative, history matching still gives a reasonably good fit and provides more information about the model structure by allowing for non-Gaussian parameter distributions. Furthermore, the true parameters are contained in the posterior distributions. We then consider history matching's ability to ingest different metrics targeting different physical parts of the model, helping to reduce parameter space further and improve model-data fit. We find the best results when history matching is used with multiple metrics; not only is the model-data fit improved, but we also gain a deeper understanding of the model and how the different parameters constrain different parts of the seasonal cycle. We conclude by discussing the potential of history matching in future studies.

How to cite: Raoult, N., Beylat, S., Salter, J., Hourdin, F., Bastrikov, V., Ottlé, C., and Peylin, P.: Exploring the Potential of History Matching for Land Surface Model Calibration, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5287, https://doi.org/10.5194/egusphere-egu24-5287, 2024.

Lake Velence is a shallow soda lake in Hungary whose water budget is mainly driven by precipitation and evaporation. The lake has shown a deteriorating tendency recently, including extremely low lake levels and poor water quality, which indicates its vulnerability against changing climatic conditions. At the same time several water usage conflicts appeared in the catchment area. Until recently, the groundwater component in the lake's water budget and the hydrogeological processes in the catchment area have not been taken into consideration. Recent hydrogeological studies, however, show groundwater discharge into the lake.  Thus, further investigating this question is of high importance, hence groundwater could reduce climatic vulnerability.

Our ongoing work aims at developing a model-based evaluation technique, utilizing all map-based geophysical information and time series of different satellite data products, having sufficient spatial resolution and providing information about parameters strongly connected to subsurface processes, showing up on the surface. The basic DEM raster layer is imported from Copernicus GLO-30 dataset, having vertical precision <4 m. The Region Of Interest is a rectangular part of the catchment area: 47.1–47.4N, 18.4–18.8E. The first segmentation of the ROI is done using elevation data combined with lithographic and soil type information, resulting in almost uniform Voronoi-like polygon tessellation, with cells classified by geostructure. Further refinement by land cover type is done using Sentinel-1 SAR data. Other fixed data of point and polygon layers are important terrain features, points of surface inflows, (known) water takeouts and monitoring wells.

The machine learning regression model has time series of measured data at all its layers, daily input from Agárd meteorological station, like precipitation, average temperature, wind speed and relative humidity. Another important input data comes from Sentinel-2 (GREEN-NIR)/(GREEN+NIR)=NDWI spectral index, available in about weekly time steps, varying between 2 days-2 weeks. A crucial feature of all remote sensing data used here is the spatial resolution being better (10 m) or similar to the resolution of the basic DEM model. During training a graph neural network is generated dynamically from the Voronoi tessellation, where cells are nodes and physical processes between neighbouring cells give edge attributes for the graph. We use rectilinear approximations for water runoff/subsurface water exchange between cells, vertical infiltration/discharge under cells and estimated evapotranspiration from them. Learnable parameters governing the intensity of these flows are connected to geostructure and land cover classes. Parameters are optimized with time interval cross validation, with one part of the time series data being left out from optimization in each epoch and used for evaluation against target water level data.

Automatic detection of spatio-temporal patterns, connected to near-surface hydrogeological processes helped visualizing and quantifying estimated physical flows. Comparison with field measurements confirmed theoretic results from MODFLOW basin modelling, proving topography as a driving factor for subsurface flows. Our model is also suitable to handle isotope tracers, and extension to deep learning model promises predictive functionality for water table level.

The research is part of a project which was funded by the National Multidisciplinary Laboratory for Climate Change, (Hungary) RRF-2.3.1-21-2022-00014.

How to cite: Rapai, T., Baják, P., Lukács, A., Székely, B., and Erőss, A.: Understanding near-surface hydrogeological processes around Lake Velence (Hungary) – using mesh graph neural networks on multidimensional remote sensing data, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5561, https://doi.org/10.5194/egusphere-egu24-5561, 2024.

In view of the growing importance of climate ensemble simulations, we propose an ensemble approach for following the dynamics of extremes in the presence of climate change. A strict analog of extreme events, a concept based on single time series and local observations, cannot be found. To study nevertheless typical properties over an ensemble, in particular if global variables are of interest, a novel, statistical approach is used, based on a ”zooming in” into the ensemble. To this end, additional sub-ensembles with initially very close members are generated around trajectories of the original ensemble. Plume diagrams initiated on the same day of a year are generated from these sub-ensembles. The trajectories within a plume diagram strongly deviate on the time scale of a few weeks. By defining the extreme deviation as the difference between the maximum and minimum values in a plume diagram, a growth rate for the extreme deviation can be extracted. An average of these taken over the original ensemble (i.e. over all sub-ensembles) characterizes the typical, exponential growth rate of extremes, and the reciprocal of this can be considered the characteristic time of the emergence of extremes. Using a climate model of intermediate complexity, these are found to be on the order of a few days, with some difference between the global mean surface temperature and pressure. Measuring the reciprocal of the growth rate in several years along the last century, results for the temperature turn out to be roughly constant, while a pronounced decaying trend is found in the last decades for the pressure.

How to cite: Jánosi, D., Herein, M., and Tél, T.: An ensemble based approach for the effect of climate change on the dynamics of extremes, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6164, https://doi.org/10.5194/egusphere-egu24-6164, 2024.

Uncertain quantities are often described through statistical samples. Can samples for numerical weather forecasts be generated dynamically? At a great expense, they can. With statistically constrained perturbations, a cloud of initial states are created and then integrated forward in time. By now, this technique has become ubiquitous in weather and climate research and operations. Ensembles are widely used, with demonstrated value.

The atmosphere evolves in a multidimensional phase space. Does a cloud of ensemble solutions encompass the evolution of the real atmosphere? Theoretically, random perturbations in high dimensional spaces have negligible projection on any direction, including the error in the best estimate, therefore consistently degrading it. As the bulk of the perturbation variance lies in the null-space of error, samples in multidimensional space do not contain reality.

An evaluation suggests that initial and short-range forecast error and ensemble perturbations are random draws from a high dimensional domain we call the subspace of possible error. Error in any initial condition is a result of stochastic observational and assimilation noise, while perturbations explore other, mostly independent directions from the subspace of possible error that may have resulted from other configurations of stochastic noise. What benefits may arise from the deterministic projection of such noise?

How to cite: Toth, Z., Feng, J., Zhang, J., and Pena, M.: A Foray of Dynamics into the Realm of Statistics: A Review of Ensemble Forecasting, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6791, https://doi.org/10.5194/egusphere-egu24-6791, 2024.

The well-posedness of the dynamic framework in earth-system model (ESM for short) is a common issue in earth sciences and mathematics. In this presentation, the authors will introduce the research history and fundamental roles of the well-posedness of the dynamic framework in the ESM,  emphasizing the three core components of ESM, i.e., the atmospheric general circulation model (AGCM for short), land-surface model (LSM for short) and oceanic general circulation model (OGCM for short) and their couplings. In fact, this system strictly obeys the conservation of energy and is used to make better climate predictions. Then, some research advances made by their own research group are outlined. Finally, future research prospects are discussed.

How to cite: Lian, R., Ma, J., and Zeng, Q.: Well-Posedness of the Dynamic Framework in Earth-System Model , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7142, https://doi.org/10.5194/egusphere-egu24-7142, 2024.

How to cite: Toppaladoddi, S. and Wells, A.: A stochastic model for the turbulent ocean heat flux under Arctic sea ice , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8590, https://doi.org/10.5194/egusphere-egu24-8590, 2024.

Numerical weather prediction and climate models encounter challenges in accurately representing flow regimes in the stably stratified atmospheric boundary layer and the transitions between them. This leads to an inadequate depiction of regime occupation statistics and, therefore, to biases in forecasts of near-surface temperature. To explore inherent uncertainties in modeling regime transitions, the response of the near-surface temperature inversion to transient small-scale phenomena is analyzed based on a stochastic modeling approach. A sensitivity analysis is conducted by augmenting a single-column model for the atmospheric boundary layer with deterministic perturbations accounting for small-scale fluctuations in the wind and temperature dynamics and with a stochastic stability function to account for turbulent bursts. The model is a tool to systematically investigate what types of unsteady flow features may trigger abrupt transitions in the mean boundary layer state. Previous research showed that incorporating enhanced mixing, a common practice in numerical weather prediction models, blurs the two regime characteristics of the stably stratified atmospheric boundary layer. Simulating intermittent turbulence through a stochastic stability function is shown to provide a potential workaround for this issue. Including key uncertainty in models could lead to a better statistical representation of the regimes in long-term climate simulation. 

How to cite: Kaiser, A., Vercauteren, N., and Krumscheid, S.: Capturing the Variability of the Nocturnal Boundary Layer through Localized Perturbation Modeling, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8958, https://doi.org/10.5194/egusphere-egu24-8958, 2024.

Understanding Earth's terrestrial biosphere dynamics is vital for comprehending our planet's environmental health and sustainability. Recently, the frequency and intensity of extreme climate events have risen, significantly impacting the biosphere. Given the advancements of recurrent neural networks in modeling complex, nonlinear dynamics, we explore the capability of recurrent neural network models to model and predict the impacts of extreme events on biosphere dynamics. In this work, we compare four different recurrent network architectures, each with distinct features: 1) Recurrent Neural Networks (RNNs); 2) Long Short-Term Memory-based networks (LSTMs), known for their efficacy in handling long-term dependencies; 3) Gated Recurrent Unit-based networks (GRUs), which offer a simplified yet effective alternative to LSTMs; and 4) Echo State Networks (ESNs), which are distinguished by fixed internal weights and training based on simple linear regression. Our study found that while recurrent network architectures show similar performance under standard conditions, Echo State Networks (ESNs) show slightly superior performance, particularly under extreme events. However, we also identify limitations in current models under extreme conditions, underscoring the need for specialized approaches to enhance predictive accuracy in these circumstances.

How to cite: Martinuzzi, F., Mahecha, M. D., Camps-Valls, G., Montero, D., Williams, T., and Mora, K.: Impact Predictability: Exploring Extremes in Biosphere Dynamics with Recurrent Neural Networks, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9569, https://doi.org/10.5194/egusphere-egu24-9569, 2024.

This study aims at proposing a new framework to perform ensemble-based estimations of dynamical trajectories of a geophysical fluid flow system. To perform efficient estimations, the ensemble members are embedded in a set of evolving reproducing kernel Hilbert spaces (RKHS) defining a family of spaces.

The method proposed here is designed to deal with very large scale systems such as oceanic or meteorological flows, where it is out of the question to explore the whole attractor, neither to run very long time simulations. Instead, we propose to learn the system locally, in phase space, from an ensemble of trajectories.

The novelty of the present work relies on the fact that the feature maps between the native space and the RKHS manifold are transported by the dynamical system. This creates, at any time, an isometry between the tangent RKHS at time t and the initial conditions. This has several important consequences. First, the kernel evaluations are constant along trajectories, instead to be attached to a system state. By doing so, a new ensemble member embedded in the RKHS manifold at the initial time can be very simply estimated at a further time. This framework displays striking properties. The Koopman and Perron-Frobenius operators on such RKHS manifold are unitary, uniformly continuous (with bounded generators) and diagonalizable. As such they can be rigorously expended in exponential forms.

This set of analytical properties enables us to provide a practical estimation of the Koopman eigenfunctions. In the proposed strategy, evaluations of these Koopman eigenfunctions at the ensemble members are exact. To perform robust estimations, the finite-time Lyapunov exponents associated with each Koopman eigenfunction (which are easily accessible on the RKHS manifold as well) are determined. On this basis, we are able to filter the kernel by removing contributions of the Koopman modes that exceed the predictability time. We show that it leads to robust estimations of new unknown trajectories. This framework allows us to write an ensemble-based data assimilation problem, where constant-in-time linear combinations coefficients between ensemble members are sought in order to estimate the QG flow based on noisy swath observations.

The methodology is demonstrated on a multilayer quasi-geostrophic model representative of the Gulf Stream area in the North Atlantic at a 10 km resolution and considering 100 training ensemble members. We show the ability of the method to estimate trajectories knowing the initial condition of a new ensemble member. Moreover, ensemble-based data assimilation is performed based on realistic swaths of altimetry observations.

How to cite: Tissot, G., Jaouen, M., and Mémin, E.: Ensemble forecasts in reproducing kernel Hilbert space family: Application on a multilayer quasi-geostrophic numerical simulation, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10075, https://doi.org/10.5194/egusphere-egu24-10075, 2024.

How to cite: Kasparian, J., Moinat, L., Gaopnenko, I., and Goyette, S.: Alternative approaches to the velocity of climate change: assessment against observations, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10174, https://doi.org/10.5194/egusphere-egu24-10174, 2024.

The forecasting of earthquake occurrence remains a significant challenge in seismology, primarily due to uncertainties in understanding the current state of stress, strength, and the parameters controlling the slip behaviour of faults. Among these parameters, friction parameters are crucial as they control the earthquake recurrence interval and their nucleation size. There is a critical gap in effectively integrating observational data with physics-based models, particularly in the face of parameter bias. This study addresses how ensemble data assimilation methods can be optimized to address these challenges and reduce uncertainties in fault-slip estimates.

Our objective is to enhance the accuracy of earthquake forecasting by incorporating model error into ensemble data assimilation methods, thus improving the estimation of critical state variables such as shear stress and slip velocity. We employed the Ensemble Kalman (EnKF), Adaptive Gaussian Mixture (AGMF), and Particle Flow (PFF) filters, which are integrated with earthquake sequence models. These methods were applied in two stages: 1) Perfect model experiments using 1-D Burridge-Knopoff models to assess the benefits of including model error in estimating non-periodic and chaotic behaviours in low-dimensional systems. 2) Application in a meter-scale direct-shear laboratory setup, assimilating measurements from shear-strain gauges near the fault, to examine the effects of varying normal stress profiles on the fault on the estimates and the impact of including model error.

The perfect model experiments demonstrated improved estimation of shear stress, slip velocity, and the state variable (θ), particularly in estimating non-periodic sequences and chaotic behaviour using stable periodic solutions when confronted with small parameter biases. In the laboratory setup, variations in normal stress profiles significantly influenced the information content. Sensor placement relative to the fault and seismic phase was found to critically impact the observations' informational value, with sensor proximity to the fault being a critical aspect and the information content being higher around the coseismic phase.

This research provides valuable insights into the intricate process of earthquake forecasting, underscoring the role of data-assimilation techniques in enhancing our understanding and forecasting abilities in this field.

How to cite: Diab-Montero, H. A., Li, M., van Dinther, Y., and Vossepoel, F. C.: Ensemble Data Assimilation with Model Error Inclusion for Estimating Fault-Slip Occurrence in Large-Scale Laboratory Experiments, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11543, https://doi.org/10.5194/egusphere-egu24-11543, 2024.

State-of-the-art data assimilation systems, such as the 4DVar system of the European Centre for Medium-Range Weather Forecasts (ECMWF), are highly successful in producing state estimates of the atmosphere constrained by millions of observations. However, existing systems require substantial approximations, e.g. in forward operators, and employ conventional models in their optimization loop. This limits the amount of information that can be extracted from observations, e.g. the assimilation of visible satellite channels is still challenging. The current approach also separates the use of observations into a data assimilation step and a forecasting one, with observations only being used indirectly for forecasting, for example for tuning of parametrizations and for evaluation.

Here, we explore the possibility to train large machine learning models for data assimilation and forecasting directly from observations. In particular, we build a generative transformer neural network that models the joint the probability distribution p(y,x) over output states y for an input x. The input are observations from a temporal window, e.g. 6h or 12h, and the output y can either be an estimate of the state within the window or a short-term forecast, e.g. for another 12h. Different observations are processed by different embedding networks but then fused in the backbone transformer network. To obtain an integrated and consistent representation of the atmospheric state that corresponds to the different input data streams, we train with a variation of the masked token model training objective from natural language processing that impels the network to learn the correlation between the different input streams and channels. To properly represent the statistical nature of the estimation of y given x, our network provides an ensemble prediction as a nonparametric model for the probability distribution over y.

We present results for a network trained with a substantial amount of data, including different satellite observations (such as AMSU-A microwave sounders from NOAA 15-19 and the METOP satellites as well as IASI), radiosondes, and ground station-based measurements. The skill for both data assimilation and forecasting is analyzed and compared to ECMWF’s operational 4DVar system. We also ablate the effect different observations have on the skill of the network output.

How to cite: Lessig, C., Lean, P., McNally, T., Alexe, M., Lang, S., Chantry, M., and Dueben, P.: Towards a machine learning model for data assimilation and forecasting directly trained from observations, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11724, https://doi.org/10.5194/egusphere-egu24-11724, 2024.

Iterative ensemble smoothers, originally developed for parameter estimation in petroleum applications, are effective data assimilation methods in coupled, unstable dynamical systems. In this study, we demonstrate this using a coupled multiscale model based on two Kuramoto-Sivashinsky equations with different spatial and temporal scales, representing two subsystems of an earth-system model. The cross-covariance between the variables of the two subsystems reflects how each subsystem influences the other, leading to unexpected structures that reveal interesting physics of the coupled system. The setup of the data assimilation allows simultaneous updating of both systems, leading to consistent estimates.

A comparative study illustrates the properties of iterative ensemble smoothers and assimilation updates over finite-length assimilation windows. We demonstrate the increased accuracy of the smoothers’ solution compared to that of the standard ensemble Kalman filter and the fast convergence of the iterations related to the efficient handling of nonlinearities by the nonlinear space-time ensemble.

Localisation, whether distance-based or adapted to spatial correlations, can be used in iterative ensemble smoothers to effectively deal with limited ensemble sizes. We discuss the effects of the spatial distribution and temporal frequency of available observations and illustrate how data gaps in one of the two subsystems affect the coupled estimate. Looking forward, we present the possibilities and benefits of a potential implementation of this approach in coupled earth-system models.

How to cite: Vossepoel, F., Evensen, G., and van Leeuwen, P. J.: Data assimilation approaches with iterative ensemble smoothers in coupled nonlinear multiscale models, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12751, https://doi.org/10.5194/egusphere-egu24-12751, 2024.

The route to chaos in a truncated rotating shallow-water model has been investigated by constructing an autonomous five-mode Galerkin truncated system with complex variables. Two distinct transitions to chaos were observed as the energy injected into the system increased. The first transition is characterized by forming a continuous sequence of bifurcations that follow the usual Feigenbaum path. The second transition, occurring for high values of injected energy, exhibits a sharp transition between quasi-periodic states and chaotic regimes. The first chaotic regime arises since nonlinear interactions are principally dominated by inertial terms, while the second one is related to the increasing importance of free surface elevation in the overall process. By rewriting the system in terms of phase and amplitude, for each variable truncated system, it has been found that phases are locked at the initial value for a certain period of time, followed by a sudden transition due to a simple rotation of $\pm \pi$, even when amplitudes show a chaotic dynamic. The time duration of phase locking decreases as the injected energy increases, and, for high values of injected energy, even phases reach a chaotic regime. This behaviour is observed since, in the nonlinear term of the equations, phases appear through linear combinations of triads of different modes. When the duration of locking periods is different for each mode, the superposition of multiple $\pi$ phases jumps, making the dynamics of the coupled phase triads stochastic, even for small values of the injected energy.

How to cite: Carbone, F. and Dutykh, D.: Route to chaos and resonant triads interaction in a truncated Rotating Nonlinear shallow-water model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-13180, https://doi.org/10.5194/egusphere-egu24-13180, 2024.

Determining the source location and release rate is critical in assessing the environmental consequences of atmospheric radionuclide releases, yet this task remains challenging due to the vast multi-dimensional solution space. To address this, we propose a spatiotemporally-separated two-step framework that reduces the dimension of the solution space in each step and enhances the reconstruction accuracy, which is applicable to radionuclide releases. The separating process involves applying a temporal sliding-window average filter to the observations, thereby reducing the influence of temporal variations in the release rate and ensuring that the features of the filtered data are dominated by the source location. Initially, candidate source locations are pre-screened using a correlation-based method. To establish the relationship between the filtered data and candidate source locations, time- and frequency-domain features are extracted from the filtered data and an eXtreme Gradient Boosting algorithm is employed for fitting. The features are further screened out by the Recursive Feature Elimination with Cross-Validation. Utilizing the features of filtered observations, the source location can be determined without the knowledge of the release rate. Subsequently, the release rate is determined using projected alternating minimization with the L1-norm and total variation regularization algorithm.

The proposed method was rigorously tested on two field experiments: the SCK-CEN experiment, featuring local-scale ⁴¹Ar releases over two days, and the ETEX-I experiment, involving continental-scale PMCH releases. Validation on the SCK-CEN experiment showed that the lowest source location error fell below 1% and the mean source location error remained under 5%, with temporal variations and peak release rate being accurately reconstructed. Similar accuracy was also observed in the ETEX-I experiment. Compared to traditional correlation-based method and Bayesian method, our method exhibited superior accuracy and a reduced uncertainty range.

Furthermore, comprehensive sensitivity tests were conducted on the SCK-CEN experiment to evaluate the influence of pre-screening range, sliding-window length, feature types, and combinations of observation sites. The results indicated that our method achieved consistent performance across various parameters and conditions, maintaining low error levels even with only a single observation site.

How to cite: Xu, Y., Fang, S., Dong, X., Hu, H., and Zhuang, S.: Spatiotemporally-separated framework for the source reconstruction of atmospheric radionuclide releases, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-13849, https://doi.org/10.5194/egusphere-egu24-13849, 2024.

The Arctic and Antarctic sea ice covers form key components of Earth's climate system. Their precipitous declines are impacting the polar marine environment and its ecosystems, with ripple effects felt far beyond the polar regions. As a material sea ice exhibits composite structure on many length scales. A principal challenge is how to use information on small scale structure to find the effective or homogenized properties on larger scales relevant to climate and ecological models. From tiny brine inclusions to rich ice pack dynamics on oceanic scales, and from microbes to polar bears, we'll consider recent advances in modeling sea ice and the ecosystems it hosts. In the spirit of MPE 2013, we’ll focus on the broad range of mathematics and physics being used. Percolation theory and statistical physics, fractal geometry, spectral analysis and random matrix theory, advection diffusion processes, topological data analysis, and uncertainty quantification for dynamical systems will arise naturally in considering various sea ice structures and organisms. This work is helping to advance how sea ice is represented in climate models, and to improve projections of the fate of Earth’s sea ice packs and the ecosystems they support.

How to cite: Golden, K.: Mathematics of Sea Ice and its Ecosystems, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-14101, https://doi.org/10.5194/egusphere-egu24-14101, 2024.

Nonlinearities in numerical models for the geosciences and in observation operators that map model states to observation space have become so strong that they can no longer be ignored. The particle flow filter (PFF) is a fully nonlinear and efficient sequential Monte Carlo filter that removes the weight degeneracy problem in particle filters by iteratively transporting the equal-weighted particles from the prior to the posterior distribution. The deterministic version of the PPF has been successfully applied to high-dimensional systems and is unbiased in the limit of an infinite number of particles. However, with a small number of particles, the ensemble spread can be biased low, especially in the observed part of the state space. This can be partly alleviated by using a so-called matrix-valued kernel in the algorithm, but the fundamental issue remains. To address this challenge, we propose a novel approach, the Stochastic Particle Flow Filter (SPFF), which includes a Gaussian noise in the Stein Variational Gradient Descent dynamics, the amplitude and covariance of which follow directly from theory. With this additional repulsive force between particles, the SPFF guarantees an unbiased posterior pdf, even with a finite number of particles.

We demonstrate the performance of the SPFF using detailed experiments with the 1000-dimensional Loreanz-96 model. Our results demonstrate that SPFF successfully avoids particle collapse of the marginal distributions and accurately captures the evolutions of particles Additionally, and initially unexpectedly, the SPFF exhibits faster convergence than the deterministic PFF and thus improves analysis accuracy compared to the PFF with a matrix-valued kernel at the computational cost. We also show results of its performance on a high-dimensional ocean model demonstrating that we, as a community, are very close to solving the nonlinear data assimilation problem.

How to cite: Yeh, H.-L. and van Leeuwen, P. J.: Unbiased fully nonlinear data assimilation: the Stochastic Particle Flow Filter, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-14208, https://doi.org/10.5194/egusphere-egu24-14208, 2024.

We consider state estimation for a system described by the one-dimensional shallow water equations. Since in general measurements of the complete state are not available, we assume that we have measurements of only one state variable, e.g. of the water height. In order to estimate the system state from these partial measurements, we construct an observer system that is based on the shallow water equations. Distributed measurements of the water height are inserted into the observer system through source terms of Luenberger type. Our main contribution is to show exponential convergence of the state of the observer system towards the original system state in the long time limit for a 2x2-system of nonlinear hyperbolic balance laws, i.e., we reconstruct the complete system state from measurements of one state variable. The proof is based on estimating the difference between the observer system and the original system via a suitable extension of the relative energy method.
Using energy-consistent coupling conditions and transforming the system to a Hamiltonian formulation, the synchronization result can be extended to star-shaped networks. This might have an application in flood modeling of river systems or control of irrigation systems.

How to cite: Kunkel, T., Giesselmann, J., and Gugat, M.: Observer-based data assimilation for shallow water equations, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15676, https://doi.org/10.5194/egusphere-egu24-15676, 2024.

Approximation, interpolation and quadrature are questions of fundamental 
importance for atmospheric and oceanic problems at planetary scale. 
Computation with spherical harmonics on the sphere is an old mathematical topic; it has a particular interest in geosciences, and is still an active field of research. In this poster, we will show numerical comparisons of several approximation schemes, with a special focus on the Cubed Sphere grid. We test hyperinterpolation, weighted least squares, and interpolation on a series of test functions with various smoothness properties. Our last results include the derivation of explicit formulas for optimal quadrature rules on low resolution Cubed Spheres.

[1] J.-B. Bellet, M. Brachet, and J.-P. Croisille, Interpolation on the Cubed Sphere with Spherical Harmonics, Numerische Mathematik, 153 (2023), pp. 249-278.
[2] J.-B. Bellet and J.-P. Croisille, Least Squares Spherical Harmonics Approximation on the Cubed Sphere, Journal of Computational and Applied Mathematics, 429 (2023),  115213.
[3] C. An and H.-N. Wu, Bypassing the quadrature exactness assumption of hyperinterpolation on the sphere, Journal of Complexity, 80 (2024), 101789.

How to cite: Croisille, J.-P., Bellet, J.-B., and Brachet, M.: Comparison of several approximation schemes on the Cubed Sphere, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15777, https://doi.org/10.5194/egusphere-egu24-15777, 2024.

Mechanistic or dynamical models based on governing equations are ubiquitous throughout science and engineering. Such models, also referred to as “simulators”, are typically characterized by a forward mapping from some set of inputs or parameters to one or more output quantities of interest. In many cases, the inputs required for the forward model are either unknown or represent approximations of unresolved processes. Bayesian inference provides a natural framework for constraining this model uncertainty using observed data. Such a framework is especially valuable in cryospheric application domains such as permafrost research, where direct observations of many quantities of interest, e.g. subsurface temperature and soil moisture, are only sparsely available. Mechanistic models based on known physics therefore play an indispensable role in filling these gaps. However, virtually all methods for Bayesian inference require repeated evaluation of the forward model which is often computationally challenging, especially for dynamical systems. As a result, computational requirements of statistical inference very quickly become intractable even for systems of only moderate complexity. The burgeoning field of “simulation-based inference” (SBI) aims to leverage modern computational methods from machine learning (ML) and data assimilation (DA) to overcome these challenges and facilitate large scale uncertainty quantification in complex scientific models. In this work, we show how SBI can be seen as a unifying theoretical framework that bridges the gap between existing DA methods (e.g. variants of the ensemble Kalman filter) and full-fledged Bayesian inference with the goal of facilitating hybrid statistical-physical modeling of complex systems. We present a novel set of software tools for making SBI more accessible to researchers along with benchmarks of several methods drawn from both the ML and DA literature. Two of these benchmarks are based on use cases from Arctic land surface modeling: degree day approximation of snowmelt and geothermal inversion of historical climate change from boreholes in Arctic permafrost. We highlight the contributions that SBI can make to solving such inverse problems and discuss further potential applications in the cryosphere and beyond.

How to cite: Groenke, B., Aalstad, K., Pirk, N., Westermann, S., Zscheischler, J., Gallego, G., and Boike, J.: Simulation-based inference as a paradigm for scientific machine learning in the cryosphere and beyond, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-16847, https://doi.org/10.5194/egusphere-egu24-16847, 2024.

Land surface models provide self-consistent estimates of the water stored in various components of the land system, i.e. in the soil, vegetation and snow, and of water fluxes. Assimilation of satellite-based data helps to update these estimates to some extent, but it has some limitations when human processes, such as irrigation, are missing in the modeling system. Furthermore, the simulated water distribution heavily depends on the choice of meteorological input and other model choices. In this study, we aim to quantify the relative contribution of (i) Sentinel-1 data assimilation for soil moisture and snow updating, (ii) meteorological input, and (iii) modeling irrigation over the Po river basin in Italy.

The Po river network channels the discharge of snow melt water from the Alps and Apennines, combined with surface and deep subsurface runoff from the hillslopes and valley. During the summer, the river network supplies irrigation water to the large agricultural area in the Po river valley. The Po basin is thus a unique testbed to study various water budget components in an environment with pronounced seasonal water storage dynamics and human water management.

More specifically, we assimilate 1-km Sentinel-1 data into the Noah-MP land surface model coupled to an irrigation module and the hydrological modeling and analysis platform (HyMAP) as runoff routing module. The Noah-MP simulations are forced with meteorological data from either the fifth generation ECMWF atmospheric reanalysis (ERA5) or the Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2) for the years 2015-2023. Sentinel-1 snow depth retrievals are assimilated over the mountains in the winter, whereas Sentinel-1 backscatter are mainly assimilated in the valley during the spring, summer and fall. The state updates are applied to snow depth, snow water equivalent, and soil moisture. These updates subsequently trigger updates in estimates of irrigation, leaf area index, discharge and other variables, resulting in a self-consistent re-analysis of the entire water budget of the Po basin. The impact of the Sentinel-1 data assimilation relative to that of the activation of irrigation modeling is quantified using independent in situ and remotely sensed measurements of soil moisture, leaf area index, snow depth, evaporation, irrigation and discharge.

How to cite: De Lannoy, G., Busschaert, L., Modanesi, S., Dunmire, D., Brangers, I., Lievens, H., Heyvaert, Z., Massari, C., Getirana, A., and Bechtold, M.: Relative contribution of high-resolution Sentinel-1 data assimilation and modeling choices to improve regional water budget estimates, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17015, https://doi.org/10.5194/egusphere-egu24-17015, 2024.

Many physical, chemical, financial, and ecological processes show the presence of a threshold, which may affect their dynamics. For instance, chemical reactions occur when the energy reaches the activation value; for insurance companies and banks, ruin may happen when the balance drops down a minimum amount; in fluvial hydraulics, sediment transport, and morphodynamic processes start when bed shear stresses (i.e., flow discharge) overcome a critical threshold. Other processes may show the presence of additional boundaries. For stochastic processes, whose dynamics depends on the frequency and magnitude of random fluctuations, it is interesting to know the average time the process takes to reach one of the critical boundaries starting from a known value. This quantity is known in the literature as Mean First Passage Times (MFPTs). The quantification of the MFPTs is usually performed by considering one threshold, only. When two or multiple thresholds are present, one may consider the MFPTs of reaching either one of the thresholds, without having passed the other ones. Such a selective condition is referred to in the literature as splitting probability. In this work, we consider stochastic processes governed by a typical Langevin equation with deterministic drift and random instantaneous jumps (white shot noise). We perform a statistical-trajectory-analysis starting from a point between two thresholds and derive exact relationships of the splitting probabilities and the MFPTs of crossing one threshold, only, based on process-dependent dimensionless parameters. Such formulations are then explicitly given for the cases of constant and linear drift functions and both positive and negative jumps. We test the derived formulations against data from MonteCarlo simulations, by varying the process parameters, the starting point, and the values of the thresholds. The comparison shows very good agreement and confirms the correctness of the derived relationships. Additionally, the analysis highlights the role played by the dimensionless parameters. Then, data from flow measurements in a river are considered and we successfully test the formulations against the duration of the raising limb of high-stage events. The derived formulations can be readily applied to calculate both the duration of the raising and the falling limbs of flow events, which are important quantities for engineering applications, as well as for modeling purposes in river eco-morphodynamics.

How to cite: Calvani, G. and Perona, P.: The selective boundary crossing in stochastic processes driven by white shot noise, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17713, https://doi.org/10.5194/egusphere-egu24-17713, 2024.

Heatwaves represent a major health hazard, as was the case for the 2003 summer heat wave, responsible for more than 70 000 deaths in western Europe. The interplay of mean flow, quasi-periodic and random fluctuations — associated with the westerly jet, Rossby and gravity waves, and eddies — of the large-scale temperature field results in complex heat wave–related amospheric conditions. Understanding how the different physical processes interact is thus crucial for prediction of heat wave events. 
In this study, we use the triple decomposition of turbulent flow (Hussain and Reynolds, JFM, 1972) to compute mean, quasi-periodic and random energy fluctuations of the temperature field, that is the 1-point energy budget of temperatures. This decomposition takes into account all interactions between the zonal jet, Rossby waves, gravity waves, and eddies. Both spectral and dynamical systems analyses are applied to the computed terms. More specifically, the concept of extremal length (Ahlfors, Vol. 371, AMS, 2010) is integrated into the equations to quantify how each term of the energy budget equations contributes to the "trapping" of temperature anomalies over Europe.      
Results show that, amid positive sea surface temperatures and negative soil moisture anomalies, during the first half of August, i.e., the hottest days of the heat wave, quasi-periodic oscillations of polar air increased, resulting in meridional migration of cold air over Canada, and subsequent mixing with warmer air coming from North America. This mixing triggered baroclinic instabilities that led to production of turbulent eddies, which by August 5th suddenly stopped their eastward progression, creating a cyclonically stalled regime over the Mid-Atlantic; this stationary cyclone interacted positively with North African warm air propagating northward over Europe, thus sustaining the dry conditions over France and much of Western Europe. The cyclonic block finally disappeared, stopping the warm air advection from North Africa, with temperatures falling just after that. 
The study reveals that interactions between quasi-periodic and random processes of production, diffusion and dissipation of a scalar field’s energy play an important role in the evolution of a major heatwave. Hence, the 2003 event was not just the result of a superposition of Rossby waves and eddy anomalies. Extremal length analysis thus reveals that the zonal advection of temperature anomalies was blocked by interactions between quasi-periodic and random production and diffusion processes. 
This study highlights the complex turbulent interactions that lead to major heat wave events, and the fact that each such event is thus unique.

How to cite: Fossa, M., Danaila, L., and Ghil, M.: Temperature at 500hPa, energy budget and the 2003 summer heatwave over Western Europe: a triple-decomposition approach, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17846, https://doi.org/10.5194/egusphere-egu24-17846, 2024.

The Kinetic Energy spectrum of the atmosphere in the mesoscales (10-500 km) is poorly understood. Aircraft measurements in the eighties first revealed that there was a kink in the spectrum, a transition from a slope of -3 to a slope of -5/3, that occurred at scales below around 400 km (Nastrom et al. [1984]). Since that time many possible mechanisms have been posited for the transition but there has been no consensus. We will present a new way of analysing the local scaling laws of geophysical data using coarse-graining, extending the work of Sadek and Aluie [2018]. Our technique allows for the creation of spatial maps of spectral slope, as well as conditioned spectra that can be used to analyse the relationship between different meteorological variables and the atmospheric kinetic energy power spectrum. This enables us to explore causes for the observed shallower slope. We observe shallower spectral slopes in regions of greater convective activity, as well as shallowing in regions of high orographic variability and interesting latitudinal effects. The important implications of our work for the celebrated Nastrom and Gage spectrum (Nastrom et al. [1984]) will be discussed.

References: 

GD Nastrom, KS Gage, and WH Jasperson. Kinetic energy spectrum of large-and mesoscale atmospheric processes. Nature, 310(5972):36–38, 1984.

Mahmoud Sadek and Hussein Aluie. Extracting the spectrum of a flow by spatial filtering. Physical Review Fluids, 3(12):124610, 2018.

How to cite: Kouhen, S., Storer, B., Aluie, H., Marshall, D., and Christensen, H.: Understanding the atmospheric kinetic energy spectrum, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17968, https://doi.org/10.5194/egusphere-egu24-17968, 2024.

Microscale pollutant dispersion is a critical aspect of air quality assessment with significant implications for the environment and public health. Designing accurate microscale dispersion models is of paramount importance for predicting air pollution exposure and assessing risks, in particular in emergency situations such as accidents at industrial sites, which are often close to densely-populated urban areas. However, this is a challenging task, as the structure and trajectory of pollutant plumes are strongly influenced by atmospheric flow, which is inherently multi-scale, turbulent, and interacts in complex ways with the built environment. To accurately capture the airflow and dispersion patterns induced by the built environment, large-eddy simulations (LES) are recognized as a high-fidelity numerical approach. However, LES are very costly and remain subject to uncertainties, partly due to the lack of knowledge and variability of the large-scale atmospheric forcing. In emergency situations, it is essential to quantify and reduce these uncertainties in order to better predict where pollutant concentration peaks occur.

In this work, to cope with the computational cost of LES, while providing the best possible information on the processes involved, we design and validate a data assimilation algorithm based on an ensemble Kalman filter (EnKF) that combines in situ concentration measurements with LES information. This LES information is obtained through a surrogate model, based on proper orthogonal decomposition (POD) combined with Gaussian process regression (GPR), which was trained in an offline stage using a large dataset of LES simulations, and which predicts the time-averaged concentration spatial fields for varying large-scale atmospheric conditions. The application of our data assimilation approach to the MUST field-scale experiment provides a proof-of-concept of the system's ability to reduce meteorological parametric uncertainties, correct bias in the model boundary conditions and thereby improve LES pollutant concentration field predictions. The use of the POD-GPR reduced-order model enables generating ensemble predictions that accurately account for the strong nonlinearities of the LES model, in just a few tens of seconds.

In addition, particular attention is paid to the representation of the errors, in particular to the internal variability of the atmospheric boundary layer that induces variability in the LES statistical fields and in the in-situ measurements. We design a bootstrap approach to quantify the significant effect of atmospheric internal variability on microscale dispersion, and we show that GPRs are able to learn this source of noise. Finally, we take internal variability into account in the data assimilation system considering that it is a source of model error and of observation error. This provides a more robust data assimilation framework with a more realistic description of the errors, which will be of interest for dispersion applications in real urban areas.

How to cite: Lumet, E., Rochoux, M., Jaravel, T., and Lacroix, S.: Surrogate-based data assimilation for microscale atmospheric pollutant dispersion, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17987, https://doi.org/10.5194/egusphere-egu24-17987, 2024.

The Atlantic Meridional Overturning Circulation (AMOC) is considered a tipping element of the earth system featuring bistability: for a given external forcing, a strong and a weak circulation state coexist as competing attracting states of the system. In the presence of random fluctuations, noise-induced transitions between the competing states are possible, posing a risk of abrupt AMOC tipping even without crossing a critical forcing threshold. It is thus crucial to better understand the stability landscape of the earth system with a multistable AMOC, particularly the properties of the boundary separating the basins of attraction of the strong and weak AMOC states. For weak noise, transitions are expected to cross the basin boundary at so-called edge states or "Melancholia states", typically chaotic saddles which are attracting on the boundary but asymptotically unstable. Here we find an edge state between the two stable AMOC states in an earth system model of intermediate complexity, PlaSim-LSG. Our approach is based on an edge-tracking technique that allows to construct a pseudo-trajectory on the chaotic saddle. We characterize the climatic and dynamical properties of this edge state and map out its location in different projections of state space. Near the edge state, the AMOC strength exhibits strong transient oscillations which we link to the ongoing physical processes. We relate our findings to the theory of unstable chaotic sets and discuss implications for the predictability of potential AMOC tipping in the future.

How to cite: Börner, R., Mehling, O., von Hardenberg, J., and Lucarini, V.: Oscillatory Melancholia state of the Atlantic Meridional Overturning Circulation in an intermediate-complexity climate model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18006, https://doi.org/10.5194/egusphere-egu24-18006, 2024.

Linking free and forced variability is one of the key challenges in climate science. As the climate system is an out of equilibrium one, the standard application of the fluctuation-dissipation theorem is out of scope. It has been shown in the past that it is possible to construct response operators that can be used to perform climate change projections using a more general formulation of response theory for nonequilibrium systems. Nonetheless, such operators lack the key property of interpretability: one cannot separate the contribution to the total response coming from different modes of natural variability of the system. We show here in a few low-dimensional models how this issue can be taken care of by taking advantage of the Koopman formalism. One can then write the response operator as a sum of terms each associated with a specific mode of variability. The obtained results also shed light on previous findings by Hasselmann and colleagues and on recently proposed data-driven methods aimed at deriving response operators from data.

How to cite: Moroney, J., Lucarini, V., and Zagli, N.: Building Response Operators Using Koopman Formalism: the Link between Free and Forced Variability, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18089, https://doi.org/10.5194/egusphere-egu24-18089, 2024.

In this work, we explore how to extend the concept of physical measures from attractors to chaotic non-attracting invariant sets. Building on Sweet and Ott’s work from 2000, we make their ideas rigorous by defining a measure on non-attracting sets in terms of Lebesgue Measure and show how to sample it numerically. We discuss its relevance for simple climate models and the sampling techniques’ limitations in the context of more complex and higher dimensional (climate) models. Knowing the measure of a non-attracting set, for example of a saddle or of an edge state (also known as melancholia state), also provides information about its fractal dimension and geometric complexity which can be useful to better understanding tipping phenomena and uncertainty close to a basin boundary.

How to cite: Roemer, R. and Ashwin, P.: Characterising Edge States: Measures on chaotic non-attracting invariant sets, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18525, https://doi.org/10.5194/egusphere-egu24-18525, 2024.

There is growing concern that various large-scale elements of the Earth system may undergo catastrophic transitions under future climate change. But already at present-day conditions there is a risk of spontaneous transitions to an undesired state induced by stochastic fluctuations, given that any of those elements occupy a multi-stable regime. For many climate sub-systems this potential present-day multi-stability is still uncertain. For instance, it cannot be ruled out that there is a regime of a collapsed Atlantic Meridional Overturning Circulation (AMOC) that is stable under present-day conditions. Assuming such an undesired stable state exists, there also exists an additional unstable state called the edge state. This state anchors the basin boundary separating the desired and undesired regimes, and it lies at the heart of the path taken by the system during a noise-induced transition between the two stable states.

In this work such an edge state lying between the stable regimes of a vigorous and a collapsed AMOC is computed for the first time in a global ocean model using an edge tracking algorithm. The physical characteristics that set this state apart from the usually observed stable regimes are analyzed. This can be useful to detect if a spontaneous collapse of the AMOC induced by stochastic climate variability is underway, or to detect so-called rate-induced tipping. It may be especially helpful if the system is close to the tipping point where the desired state loses stability. Here the desired stable state and the edge state become increasingly similar, but a transition towards the undesired state may nevertheless be detected early-on if specific signatures of the edge state are recognized.

How to cite: Lohmann, J.: Characterization of Edge States as Gateway to a Collapse of the Atlantic Ocean Circulation, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18809, https://doi.org/10.5194/egusphere-egu24-18809, 2024.

Environmental and hydrological models have become important decision-making tools. The PESHMELBA model [1] is a pesticide transfer model used to simulate and compare possible land-use planning scenarios in order to identify developments that reduce the impact of pesticides in surface waters.

When calibrating environmental models, one of the first steps is a sensitivity analysis [2]. However, this analysis can vary for different realizations of the external conditions (such as rainfall, evapotranspiration, or date of pesticide application) under which the model operates. Indeed, even the external conditions as an uncontrollable stochastic quantity means that the hydrological model itself becomes stochastic. Sobol indices can then be seen as random variables [3], where randomness is given by their dependence on rainfall.

In this study, we calculate the sensitivity indices on two examples: first, on soil hydrodynamic parameters on a vineyard plot in PESHMELBA under several realizations of the same type of rainfall, corresponding to events measured on the field. Second, on a omore complex situation considering the whole watershed and pesticides output variables. The stochasticity of that second case comes from the difference between the rainfall and the date of pesticide application, whicih is a key unknown in pesticide transfer simulations.

We show that the hierarchy of input parameters varies according to the forcings used. In particular, heavier rainfall mainly activates processes in the deep saturated horizon, involving parameters governing saturated soil properties (water content at saturation, for example), which is not the case for lighter rainfall, for which PESHMELBA is essentially influenced by unsaturated soil parameters.

The aim of this work is to take this dependency into account within the sensitivity analysis and to propose a global indice which is valid condidering the forcings uncertainties.

[1] Rouzies, E., Lauvernet, C., Barachet, C., Morel, T., Branger, F., Braud, I., & Carluer, N. (2019). From agricultural catchment to management scenarios: A modular tool to assess effects of landscape features on water and pesticide behavior. Science of The Total Environment, 671, 1144–1160. https://doi.org/10.1016/j.scitotenv.2019.03.060

[2] Mai, J. (2023). Ten strategies towards successful calibration of environmental models. Journal of Hydrology, 620, 129414. https://doi.org/10.1016/j.jhydrol.2023.129414

[3] Hart, J. L., Alexanderian, A., & Gremaud, P. A. (2017). Efficient Computation of Sobol' Indices for Stochastic Models. SIAM Journal on Scientific Computing, 39(4), A1514–A1530. https://doi.org/10.1137/16M106193X

How to cite: Lauvernet, C., Radisic, K., and Vidard, A.: Sensitivity analysis with external stochastic forcings for robust calibration : application to a water and pesticide transfer model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-19838, https://doi.org/10.5194/egusphere-egu24-19838, 2024.

In this study, we introduce a variety of ocean eddy parameterizations and discuss how they affect the representation of mesoscale turbulence in ocean models. They ultimately all aim at reducing overdissipation at eddy-permitting resolutions by utilizing the inverse energy cascade and energy conversions between potential and kinetic energy.

Mesoscale eddies play a crucial role in the global oceans. They transport tracers and heat, cascade energy across scales and interact with the mean currents and the atmosphere. However, their representation at resolutions close to the Rossby radius of deformation is insufficient. Such eddy-permitting, i.e. barely eddy resolving grids are still commonly applied for decadal climate simulations and will remain state-of-the-art at high latitudes for years to come. These simulations generally suffer from an excessive dissipation of kinetic energy, leading to reduced eddy variability, eddy formation and eddy-mean flow interactions.

Reducing overdissipation via optimized viscous closures is one way forward. Another option is to reinject some of the overdissipated energy back into the resolved flow via so called kinetic energy backscatter parameterizations. We will investigate different methods how to complement our own viscous and backscatter schemes with stochastic components, to account for unresolved chaotic variations of dissipative processes and for scale interactions across the resolution limit. For this purpose, we use data informed approaches such as linear inverse models to generate stochastic patterns based on high resolution reference simulations of idealized channel and double gyre configurations. Our results show that incorporation of such schemes can help to substantially improve the kinetic energy and mean flow characteristics. Furthermore, the varied application of the noise can reveal pathways of energy conversion between potential and kinetic energy, shedding light on the simulated energy cascades at such model resolutions. Aside from the learned construction of the stochastic patterns based on high resolution data, these new schemes come at a small additional computational cost, especially compared to higher resolution simulations. When tuned with caution, they provide a means to incorporate model uncertainty and to reduce systematic biases in ocean models.

How to cite: Juricke, S., Bagaeva, E., Danilov, S., Franzke, C., and Oliver, M.: Ocean eddy parameterizations: Stochastic and deterministic approaches for kinetic energy injection, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-19974, https://doi.org/10.5194/egusphere-egu24-19974, 2024.

Low-order climate models have played an important role in understanding the low-frequency variability of the atmospheric circulation and how it can be affected by trends in anthropogenic forcing. A simple quasi-geostrophic model of the midlatitudes’ circulation (Lorenz, Tellus, 1984, 1990) is studied from the perspective of the theory of nonautonomous dynamical systems (NDS: e.g., Ghil et al., Physica D, 2008).

We start with a study of the model’s behavior in the absence of time dependent forcing and determine in this case its steady states. A bifurcation analysis is carried out in order to identify distinct regime behavior types – stationary, periodic and chaotic – in the model’s parameter space. Next, we study the nonautonomous system with a meridional temperature gradient that varies seasonally, according to changes in insolation. The snapshot attractor (Tel et al., JSP, 2020) of the seasonally forced model is compared with the standard forward attractor of the autonomous model for two distinct epochs of the year, at peak summer and peak winter. In both cases, the effects of the change in forcing are reflected in a clear change of shape of the attractor. Predictability is lost in both cases: the summer attractor loses its periodicity when the forcing is seasonal. The winter one favors energy transport through one of the two wave components included in the model. For the same value of the forcing, the structure of the attractor in the autonomous case is quite different from that in the nonautonomous one.

Finally, the meridional forcing is subjected to climate trends, both positive and negative, since the jet intensity changes in opposite directions at low and high altitudes (Lee et al., Nature, 2019). The analysis of the snapshot attractor of the system under climate trends suggests that the model does not follow the geostrophic assumption in certain ranges of the forcing, as the average zonal flow does not always show the expected dependence on the equator-to-pole temperature contrast. On the other hand, the energy transported by the eddies does follow the sign of the climatic trend. Overall, distinct effects are observed. Chaotic behavior can be completely suppressed in favor of a regularly periodic one and vice-versa. At the same time, circulation patterns can change, suddenly disappear, and be restored.

In general, the snapshot attractor proved to be a robust tool in studying the internal variability of the midlatitude circulation, as well as the changes arising in it from anthropogenic forcing trends. The distinct regimes of behavior are being examined more closely by advanced spectral analysis methods (Ghil et al., Rev. Geophys., 2002) to better understand the effects of climate trends on low-frequency variability.

How to cite: Maraldi, B., Dijkstra, H., and Ghil, M.: Intraseasonal atmospheric variability under climate trends, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-21689, https://doi.org/10.5194/egusphere-egu24-21689, 2024.

How to cite: Morr, A., Riechers, K., Rydin Gorjão, L., and Boers, N.: Anticipating critical transitions in multi-dimensional systems driven by time- and state-dependent noise, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1852, https://doi.org/10.5194/egusphere-egu24-1852, 2024.

The Atlantic Meridional Overturning Circulation (AMOC) is projected to weaken due to the increase in buoyancy caused by anthropogenic warming and consequent freshening during the 21^st century and beyond. Atmosphere-ocean general circulation models simulate this AMOC weakening due to warming of the surface ocean, changes in the hydrological cycle that shift the North Atlantic salt budget, melting sea-ice, and changes in the atmospheric circulation. However, the freshwater contribution from the melting Greenland ice sheet is often either only considered in idealized scenarios or entirely omitted due to computational constraints. This simplification contributes to the large uncertainty surrounding the possibility of the AMOC crossing a tipping point in the forthcoming centuries. Here we employ the fully coupled Earth system model of intermediate complexity Bern3D v3, which dynamically simulates all ice-ocean-atmosphere interactions. We conduct a set of simulations driven by idealized CO₂ concentration paths to investigate the impact of the melting Greenland ice sheet on the stability of the AMOC over the next 3000 years. We find that for a slow CO₂ increase of 0.5%/yr up to twice pre-industrial levels, the general trends of the AMOC evolution are independent of whether Greenland meltwater is taken into account, with an initial weakening, but long-term recovery. Yet, the additional meltwater results in a further weakening of about 3 Sv after 100 years, but without leading to a full collapse of the circulation. This effect is due to melt rates remaining relatively low for the initial 100 years and only reaching their peak after 500 years. In the long-term, the curtailed AMOC and hence northward heat transport substantially slows down the disintegration of the Greenland ice sheet. Only in scenarios where the melt rates are kept artificially high, the AMOC does not recover. This highlights that the meltwater-induced AMOC weakening stabilizes the Greenland ice sheet, which in turn limits further AMOC weakening. This suggests that the potential for cascading interactions may be limited.

How to cite: Pöppelmeier, F. and Stocker, T. F.: Impact of future Greenland ice sheet melt on the stability of the Atlantic Meridional Overturning Circulation, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3874, https://doi.org/10.5194/egusphere-egu24-3874, 2024.

The present-day Atlantic Meridional Overturning Circulation (AMOC) is considered to be a prominent tipping element and its collapse would have grave consequences on the global climate. Hence, it is important to determine probabilities and pathways for noise-induced tipping events. However, as there is no observational evidence for an AMOC transition over the historical period, a noise-induced transition is expected to be a rare event in models and simple Monte Carlo techniques are not suited for such low-probability events. Here, we use Large Deviation Theory to directly compute the most probable transition pathways for the collapse and recovery of the AMOC in a box model of the World Ocean calibrated to the FAMOUS-model, where we added stochastic freshwater forcing. This allows us to determine the physical mechanisms of noise-induced AMOC transitions. We show that the most likely path of an AMOC collapse starts paradoxically with a strengthening of the AMOC followed by an immediate drop within a couple of years due to a short but relatively strong freshwater pulse. The recovery on the other hand is a slow process, where the North Atlantic Ocean needs to be gradually salinified over a course of decades, and its dynamics are quite close to the recovery in a bifurcation tipping event. The proposed method provides several benefits, including an estimate of probability ratios of collapse between various freshwater noise scenarios, showing that the AMOC is most vulnerable to freshwater forcing into the Atlantic thermocline region.

How to cite: Soons, J., Grafke, T., and Dijkstra, H. A.: Optimal Transition Paths for AMOC Collapse and Recovery in a Stochastic Box Model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4021, https://doi.org/10.5194/egusphere-egu24-4021, 2024.

Observations are increasingly used to detect critical slowing down (CSD) in potentially multistable components of the Earth system in order to warn of forthcoming critical transitions in these components. In addition, it has been suggested to use the statistical changes in these historical observations to extrapolate into the future and predict the tipping time. We argue that this extrapolation is too sensitive to uncertainties to give robust results. In particular, we raise concerns regarding (1) the modelling assumptions underlying the approaches to extrapolate results obtained from analyzing historical data into the future, (2) the representativeness of individual time series representing the variability of the respective Earth system components, and (3) the effect of uncertainties and preprocessing of the employed observational datasets, with focus on non-stationary observational coverage and the way gaps are filled. We explore these uncertainties both qualitatively and quantitatively for the Atlantic Meridional Overturning Circulation (AMOC). We argue that even under the assumption that these natural systems have a tipping point that they are getting closer to, the different uncertainties are too large to be able to estimate the time of tipping based on extrapolation from historical data.

How to cite: Ben Yami, M., Morr, A., Bathiany, S., and Boers, N.: Uncertainties too large to predict tipping times of major Earth system components, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5870, https://doi.org/10.5194/egusphere-egu24-5870, 2024.

Recent studies utilizing the CMIP5 and CMIP6 model ensembles reveal that the subpolar North Atlantic (NA) is prone to deep convection collapsing leading to abrupt cooling of sea surface temperatures. Consequently, the latest comprehensive study on tipping points and the first report on global tipping points include the subpolar gyre (SPG) deep convection on the list of core tipping elements of Earth’s climate system.

Here, we investigate the drivers and impacts of a collapse of deep convection in the subpolar NA and the role of internal variability using a coupled climate model large ensemble (namely, the CESM2-LE consisting of 100 ensemble members) under the SSP3-7.0 forcing scenario. We identify that freshening of surface conditions leads to the negative surface density anomaly, eventually resulting in the cessation of deep mixing and the abrupt cooling of sea surface temperatures. The ensemble shows abrupt cooling occurring approximately in 2045 with internal variability leading to a spread of ±11 years. In each ensemble member, the subpolar NA transitions to a new state without deep convection, colder sea surface temperatures, strongly reduced heat loss to the atmosphere, and large circulation changes.

Internal variability does not determine if, but when abrupt cooling occurs, suggesting a forced response to larger-scale changes and a potential tipping point to be reached decades before the prominent abrupt cooling event. We provide evidence for the collapse of deep convection being a component of a positive feedback mechanism resulting in the SPG circulation transitioning to a weaker state. Without deep convection at the center of the circulation, the density gradient-driven part of the gyre circulation vanishes and the circulation strength decreases by approximately 50 %. The tipping point of the subpolar NA is therefore reached decades prior to the abrupt cooling and abrupt cooling is an inevitable consequence of the tipping event.

This points towards a potential misconception concerning drivers of abrupt climate
change in the subpolar NA, connected tipping points, and their thresholds, highlighting
the necessity for clarifying research efforts in the future.

How to cite: Müller, J., Zappa, G., and Bellucci, A.: North Atlantic Subpolar Gyre Deep Convection: A Tipping Point Reached Decades Ago?, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5998, https://doi.org/10.5194/egusphere-egu24-5998, 2024.

Abrupt transitions in ecosystems can be interconnected, raising challenges for science and management in identifying sufficient interventions to prevent them or recover from undesirable shifts. Here we use principles of network controllability to explore how difficult it is to manage coupled regime shifts. We find that coupled regime shifts are easier to manage when they share drivers, but can become harder to manage if new feedbacks are formed when coupled. Simulation experiments showed that both network structure and coupling strength matter in our ability to manage interconnected systems. This theoretical insights calls for an empirical assessment of cascading regime shifts in ecosystems and warns about our limited ability to control cascading effects.

How to cite: Rocha, J. and Crépin, A.-S.: Structural controllability and management of cascading regime shifts, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7515, https://doi.org/10.5194/egusphere-egu24-7515, 2024.

To tip or not to tip? Many subsystems of the Earth are at risk of undergoing abrupt transitions from their current stable state to a drastically different and often less desired state due to anthropogenic climate change. These so-called tipping events often present severe consequences for ecosystems and human livelihood that are difficult to reverse. One common mechanism for tipping to occur is via forcing and driving a nonlinear system beyond a critical threshold that signifies self-amplifying feedbacks inducing tipping. However, previous work has shown that it is possible to briefly overshoot a critical threshold and avoid tipping. Specifically, the peak distance of an overshoot and the time a system can spend beyond a threshold are governed by an inverse square law relationship. In the real world or complex models, critical thresholds and other system features determining the permitted overshoot are highly uncertain. In this presentation, we look at how such uncertainties affect the probability of tipping from the perspective of uncertainty quantification. We show the importance of constraining uncertainty in the location of the critical threshold and the linear restoring rate to the stable state to reduce the uncertainty in the probability of tipping. Using a simple box model for the Atlantic Meridional Overturning Circulation, we highlight the need to constrain the high uncertainty found in wind-driven fluxes represented by a diffusive time scale within the box model to reduce uncertainty in the tipping probability for overshoot scenarios. 

How to cite: Ritchie, P. and Lux-Gottschalk, K.: Uncertainty quantification for overshoots of tipping thresholds, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8618, https://doi.org/10.5194/egusphere-egu24-8618, 2024.

Terrestrial ecosystems are affected by climate change, deforestation and other human influences. There is concern that the resilience of these ecosystems, i.e. their ability to recover from perturbations, is thereby decreased and that their sensitivity to environmental change is increased. In the extreme case, this sensitivity could diverge at a “tipping point”, and propel systems into alternative states. A prominent example is the potential dieback of the Amazon rainforest and the transition to a savanna-like state.

The notion of resilience is a highly complex and multi-faceted concept. Ecological resilience theory and the mathematical properties of dynamical systems suggest that a number of different resilience quantifiers are related to each other, or even equivalent, which would allow improved “resilience monitoring” from space. For instance, indicators based on the phenomenon of “critical slowing down” (CSD) like variance and autocorrelation, and related indicators have been used to detect changes over time. In contrast to empirical recovery rates, these indicators do not require one to directly observe the recovery from rare extreme disturbances. Also, they do not rely on the observation or attribution of the responsible environmental drivers.

Based on the assumption that fluctuations in remotely sensed proxies of vegetation properties (like biomass or vegetation greenness) behave like iconic one-dimensional stochastic models (most importantly, the Ornstein-Uhlenbeck process), CSD-based indicators should be related to empirical recovery rates after perturbations, to the more general Kramers-Moyal coefficients rooted in statistical mechanics, and to the sensitivity of a dynamical equilibrium state to environmental change. It has been shown that in observations, the theoretically expected relationships between some of these measures roughly hold. At the same time, process-based models, as well as observations, can deviate from such simple stochastic models, e.g. when multiple plant types affect the resilience of an ecosystem but not its sensitivity to environmental change.

In our contribution, we show and discuss examples for such deviations in a global vegetation model LPJ. In addition, we compare resilience indicators across a number of state-of-the-art models from CMIP6 and compare the results to an assessment of observations, in order to separate limitations that are related to the practical measurement process (e.g. uncertainties related to retrieval algorithms) from limitations that are associated with unjustified theoretical assumptions. Our results are meant to guide resilience monitoring toward meaningful indicators and to focus on regions and observable properties that can warn of future loss of ecosystem services.

How to cite: Bathiany, S., Blaschke, L., Morr, A., and Boers, N.: Consistency of resilience indicators in terrestrial vegetation models, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8927, https://doi.org/10.5194/egusphere-egu24-8927, 2024.

Recent quasi-equilibrium studies performed in the Community Earth System Model (CESM) have revealed a bi-stable regime of the Atlantic Meridional Overturning Circulation (AMOC) in this model. This suggests that the present-day AMOC might exist in a bi-stable regime, emphasizing the need for accurate predictions regarding the probability of an AMOC collapse over the next decades. However, the CESM exhibits notable biases, with a critical freshwater transport bias at 34°S in the Atlantic emerging as a key determinant of AMOC stability. Specifically, this bias enhances the stability of the AMOC, rendering the CESM unable to accurately predict the likelihood of AMOC tipping.

In this study, we establish a direct connection between the freshwater transport bias in the CESM and a corresponding freshwater content bias in the Indian Ocean. By investigating the detailed freshwater balance, we identify specific regions within the Indian Ocean that exert a significant influence on the Atlantic freshwater transport bias at 34°S. This quantitative analysis enables us to construct an optimal surface-flux correction, which reduces the model biases. This physics-based surface-flux correction allows us to adjust the AMOC to its correct stability regime in the CESM without imposing unrealistic flux adjustments

How to cite: Vanderborght, E., Dijkstra, H., and Westen, R.: Origin in of the AMOC fresh water transport biases in a state-of-the-art climate model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11804, https://doi.org/10.5194/egusphere-egu24-11804, 2024.

 In a recent paper [2] we predicted a collapse of the AMOC as soon as mid-century at odds with assessments based on climate model scenarios. The prediction was based on the sub polar gyre fingerprint as a proxy for the AMOC as proposed by Ceasar et al. [2]. Several other fingerprints have been proposed, all showing early warning signals of a forthcoming tipping point [3]. Here we present a statistical analysis, optimally extracting the common signal in the different fingerprints in order to further solidify the assessments. 

[1] Ditlevsen, P., Ditlevsen, S. Warning of a forthcoming collapse of the Atlantic meridional overturning circulation. Nat Commun 14, 4254 (2023)

[2] Caesar, L., Rahmstorf, S., Robinson, A. et al. Observed fingerprint of a weakening Atlantic Ocean overturning circulation. Nature 556, 191–196 (2018)

[3] Boers, N. Observation-based early-warning signals for a collapse of the Atlantic Meridional Overturning Circulation. Nat. Clim. Chang. 11, 680–688 (2021)

How to cite: Ditlevsen, P.: Fingerprinting the AMOC and predicting a collapse, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12023, https://doi.org/10.5194/egusphere-egu24-12023, 2024.

Rate-induced tipping (R-tipping) occurs when the forcing rate changes too rapidly for the system to track its quasi-equilibrium state, leading to an unexpected collapse. Currently, there is a lack of valid early warning signals (EWS) for R-tipping, particularly in the presence of noise perturbations. To address this deficiency, we employ a deep learning algorithm to extract the high-order structures hidden within time series data before R-tipping occurs. Then the trained neural networks are taken to provide real-time EWS for R-tipping, demonstrating skillful forecasts with a substantially long lead time, surpassing the performance of conventional critical slowing down indicators. Our progress underscores the predictability of R-tipping, offering the potential to improve the ability to deduce the safe operating space for a wider spectrum of complex systems.

How to cite: Huang, Y., Bathiany, S., Ashwin, P., and Boers, N.: Anticipating rate-induced tipping by a deep learning framework, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12187, https://doi.org/10.5194/egusphere-egu24-12187, 2024.

Many potential tipping elements have been identified in the climate system over the last decade, although some of them are surrounded by large uncertainties. We perform an updated analysis of abrupt changes in current state-of-the-art climate models to re-evaluate the evidence of these shifts—whether they are tipping points or not. We examine all CMIP6 models (59 in total) under the 1pctCO2 scenario using a Canny edge detection method—adapted for spatiotemporal dimensions—to detect abrupt shifts in climate data. We perform this semi-automatic analysis on 83 two-dimensional variables of the ocean, atmosphere, and land. We aggregate the detected shifts that are connected spatially or temporally. This results in connected regions of abrupt shifts and allows us to map areas that are most at risk of these shifts according to CMIP6 models. We report statistics on number of abrupt changes detected, surface area of abrupt changes, and critical global mean temperature at which these abrupt changes occur. This is done for various climate subsystems and potential tipping elements, such as the Arctic sea ice, Antarctic sea ice and the North Atlantic subpolar gyre. We find evidence for abrupt changes in several systems, but not all models show them equally.

How to cite: Terpstra, S., Falkena, S. K. J., Bastiaansen, R., Bathiany, S., Dijkstra, H. A., and von der Heydt, A.: Analysis of Abrupt Changes in CMIP6 Models Using Edge Detection, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12295, https://doi.org/10.5194/egusphere-egu24-12295, 2024.

Early Warning Signals (EWS) are indicators that can be used to anticipate tipping points i.e. abrupt changes in dynamical systems. Detecting EWS is a crucial part of climate science, especially in the context of climate change. Several methods are used to identify tipping points using time series of climate state variables (e.g. temperature, precipitation, etc), but few consider spatial correlations [1]. Spatial detection could identify the starting location of a transition process from a state to another and be directly applied to satellite observations. We consider different state variables on a numerical grid as a complex network, where grid points displaying correlation are connected and the temporal evolution of this network is studied. 
We seek for network properties that can be used as EWS when approaching the state transition. 

The network is generated and analysed using the pyUnicorn package [2], and compared to classical statistical methods. The networks are constructed using two methods: Pearson correlation coefficient and mutual information, allowing us to compare a linear and a causal approach. Multiple network indicators such as the degree of correlation, the average path length, and the area weighted connectivity are compared. To test the method robustness, we look at the network dependencies in terms of the time window, the interval over which the forcing is changed, and the effect of reducing the extent of the network (limited, for example, over polar or equatorial regions). These indicators show tipping points at the global scale, as simulated in a coupled-aquaplanet configuration with the MIT general circulation model, using as forcing parameter the atmospheric CO₂ content or the input of solar energy [3] . The application of such indicators as EWS is discussed.

[1] van der Mheen et al. Geophysical Research Letters 40, 11  (2013)

[2] Donges et al. Chaos 25, 113101 (2015)

[3] Brunetti \& Ragon, Physical Review E 107, 054214 (2023)

How to cite: Moinat, L., Kasparian, J., and Brunetti, M.: Investigating Early Warning Signals in Climate Simulations using Complex Networks, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-16007, https://doi.org/10.5194/egusphere-egu24-16007, 2024.

Tropical forests are vital for climate change mitigation as carbon sinks. Yet, research suggests that climate change, deforestation and other human influences threaten these systems, potentially pushing them across a tipping point where the tropical vegetation might collapse into a low-treecover state. Signs for this trend are reductions of resilience defined as the system's capability to recover from perturbations. When resilience decreases, according to dynamic system theory, a critical slowing down (CSD) induces changes in statistical measures such as the variance and the autocorrelation. This allows to indirectly examine resilience changes in the absence of observations of strong perturbations. Yet, deriving estimates of the statistical measures indicating resilience changes based on CSD impose several assumptions on the system under observation. For tropical vegetation, it is not obvious that these assumptions are fulfilled.

Additionally, the conditions of tropical rainforests pose difficulties on the observation of the vegetation. Among other factors, cloud cover, aerosols, and the dense vegetation hinder the reliable retrieval of Vegetation Indices (Vis), especially from data gathered in the optical spectrum. Thus, such data might not be suitable for resilience analyses based on CSD, even if the theory is applicable in principle.

We investigate the different assumptions of CSD and test them on a diverse set of remotely sensed VIs. Hereby, we establish a framework that allows to decide whether a specific dataset is appropriate for resilience analyses based on CSD.

How to cite: Blaschke, L., Morr, A., Bathiany, S., Telschow, F., Smith, T., and Boers, N.: Applicability of CSD-based resilience analyses to remotely sensed Vegetation Indices in the Tropics, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17709, https://doi.org/10.5194/egusphere-egu24-17709, 2024.

The Tipping Point Modelling Intercomparison Project (TIPMIP) is an international initiative that aims to systematically improve our understanding of potential tipping dynamics in different components of the Earth system and to assess the associated uncertainties (www.tipmip.org). By linking and evaluating different models through a systematic framework, TIPMIP aims to fill critical knowledge gaps in Earth system and climate risks by improving their assessment at different levels of anthropogenic forcing and associated long-term commitments and irreversibilities. The Methods and Risk Assessment Working Group of TIPMIP will further develop the methodological foundations of this systematic approach to the study of tipping dynamics in domain-specific and coupled Earth system  numerical models to support future assessment reports and comprehensive risk analyses. In this contribution, we introduce the Methods and Risk Assessment Working Group within TIPMIP, and highlight relevant lines of methodological development to be pursued, including: (i) systematic and automated detection of tipping points and critical transitions in model output and Earth observation data for TIPMIP (e.g, the TOAD framework), (ii) detection of non-linear regime shifts in time series data for TIPMIP beyond amplitude shifts, e.g. transitions between more regular and more erratic variability (e.g. the pyunicorn toolkit), and (iii) probabilistic analysis and emulator approaches of risks for triggering tipping events and cascading tipping dynamics at different levels of anthropogenic forcing (e.g. the pycascades approach).

How to cite: Donges, J. F., Dennis, D. P., Loriani, S., Sakschewski, B., Wunderling, N., and Winkelmann, R.: Methodologies for climate tipping points analysis and risk assessments in TIPMIP, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18039, https://doi.org/10.5194/egusphere-egu24-18039, 2024.

A tipping cascade is a series of tipping events in the Earth system where transitions in one subsystem can trigger further transitions in other subsystems. A concern for the future is that such a cascade could lock the Earth system in a pathway towards a so-called hothouse state. We investigate this possibility with SURFER, a reduced complexity model with a process-based carbon cycle that can reliably predict CO₂ concentrations, global mean temperatures, sea-level rise, and many ocean acidification metrics on timescales from decades to millions of years. We have incorporated in the model a network of interacting tipping elements and their feedback on the climate through albedo changes and additional greenhouse gas emissions. This has allowed for a systematic investigation of the effects of a family of realistic emission scenarios on the future trajectories of the Earth system. Our results show that a permanent shift to a hothouse state within the next few centuries is implausible. On longer time scales, however, tipping cascades can lead to enduring additional warming and particularly sea level rise.

How to cite: Couplet, V., Martínez Montero, M., and Crucifix, M.: Tipping cascades, future Earth system trajectories and the prospect of a hothouse: insights from the SURFER model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18045, https://doi.org/10.5194/egusphere-egu24-18045, 2024.

The Earth's climate system is a complex system that includes key components such as the Arctic Summer Sea Ice or the El Niño Southern Oscillation as well as climate tipping elements like the continental-scale ice sheets or the Amazon rainforest. Crossing warming thresholds of these elements can lead to a qualitatively different climate state, endangering the stability of human societies. Particularly, the cryosphere elements are vulnerable at current levels of global warming (1.2°C) while also having long response times and large structural uncertainties. Investigating a network of interacting Earth system components using an established conceptual model, we systematically assess which uncertainties of key Earth system component have the largest impacts on tipping risks. We find that the cryosphere tipping elements (the Greenland and the West Antarctica ice sheets) are most decisive for tipping risks and cascading effects within our model. At a global warming level of 1.5°C, neglecting the large cryosphere tipping elements can reduce the mean number of disintegrated Earth system components by as much as 56%. This is concerning as overshooting 1.5°C of global warming is fast becoming inevitable, while current state-of-the-art IPCC-type global circulation models do not (yet) include dynamic ice sheets. Our results suggest that urgent integrated Earth system model development and Earth observation efforts including the large polar ice sheets are necessary and a precautionary measure of meeting stringent climate targets is crucial to limit tipping risks.

How to cite: Rosser, J., Winkelmann, R., and Wunderling, N.: Cryosphere tipping elements decisive for tipping risks and cascading effects in the Earth system, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18126, https://doi.org/10.5194/egusphere-egu24-18126, 2024.

Extreme cold winters have been projected to decrease in the future, although their impacts on society are still significant. The goal of this study is to assess whether climate change affects the atmospheric mechanisms leading to cold winters.

We first explore the dynamics of 15-day winter cold spells in France, as observed since 1950. We find that the most extreme events tend to have the same atmospheric circulation pattern, consisting of an eastward-shifted NAO- dipole. We calculate an atmospheric index that characterizes this dipole. Then, using a stochastic weather generator with importance sampling, we show that this is a sufficient condition to trigger extremely cold temperatures in France, and that it performs better than a classical North Atlantic Oscillation index. This suggests that a dipole of atmospheric circulation is a necessary and sufficient condition leading to extreme cold spells in France.

We use this atmospheric index to select the CMIP6 models that best reproduce the identified dynamics leading to extreme cold spells of 15 days. Using a stochastic weather generator with importance sampling, we run simulations of worst-case winter cold spells from 2015 to 2100, following different emission trajectories for the selected models. 15-day winter cold spells in France will reach less extreme temperatures at the end of the century, especially in the case of a high-emission scenario (SSP5-8.5). However, the simulated ensembles of extreme cold spells do not show the same warming trend as the mean temperature, and very extreme cold spells are still possible in the near future. The atmospheric circulation prevailing during these events is analyzed and compared with the circulation observed during previous events.

How to cite: Cadiou, C. and Yiou, P.: Assessing changes in the intensity and dynamics of extreme cold spells in France from CMIP6, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-622, https://doi.org/10.5194/egusphere-egu24-622, 2024.

Climate models aim at representing as closely as possible the statistical properties of the climate components, including extreme events on which fine-tuning data may be less available. This is a fundamental requirement to correctly project changes in their dynamics due to anthropogenic forcing.
In order to evaluate how closely models match observations, we need algorithms capable of selecting, processing and evaluating relevant dynamical features of the climate components. This has to be reiterated efficiently for large datasets such as those issued from the Coupled Model Intercomparison Project 6 (CMIP6). In this work, we use Latent Dirichlet Allocation (LDA), a statistical soft clustering method initially designed for natural language processing, to extract synoptic patterns from sea-level pressure data and evaluate how close the dynamics of CMIP6 climate models are to ERA5 reanalysis, both in the general case and in the case of extreme temperature events.

LDA allows for learning a basis of decomposition of maps into objects called "motifs". From the ERA5 sea-level pressure data, the method robustly extracts a basis of motifs that are interpretable objects at synoptic scale, i.e. cyclones or anticyclones. Pressure data can be projected onto this basis, yielding motif weights that contain local information about the large-scale atmospheric circulation. LDA decomposition is efficient and sparse: most of the information of a given map is contained in few motifs. It is therefore possible to decompose any map in a limited number of easy-to-interpret synoptic objects. This allows for a variety of new angles for statistical analysis.

The weights statistics can be used to characterize the general and extreme dynamics in reanalysis and model data. By comparing the statistics obtained from reanalysis data with those obtained from a selection of CMIP6 models, we can quantify errors on each localized circulation pattern and identify model-specific and model-agnostic errors. We found that, on average, large-scale circulation is well predicted by the models, but model errors are increased for extreme events such as heatwaves and cold spells. Additionally, Mediterranean motifs were found to be associated with significant model errors for all the considered models in all cases.

How to cite: Malhomme, N., Podvin, B., Faranda, D., and Mathelin, L.: Evaluation of atmospheric circulation of CMIP6 models for extreme temperature events using Latent Dirichlet Allocation, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1181, https://doi.org/10.5194/egusphere-egu24-1181, 2024.

The European summer of 2022 has been widely recognized as the warmest since mid-19^th century. Our updated analyses of instrumental and reconstructed temperature series since 1500 indicate that the European summer (June-to-August) of 2022 was the warmest on record, exceeding the previous hottest summer of 2021 by a large margin. In fact, the past three summers of 2021–2023 have been among the hottest ones of the last five centuries.

By applying a heatwave (HW) detection algorithm to reanalysis data, we identify three large European HW events that affected ample regions of the continent in mid-June, mid-July and August/early September 2022. These episodes were triggered by high-pressure systems with noticeable differences in their characteristics. Additional analyses confirm that high-latitude blocks were largely responsible for the August 2022 HW, whereas subtropical ridges dominated during the June and July 2022 HWs. These HW events were also accompanied by dry soils and warm Sea Surface Temperatures (SSTs) over the Mediterranean. Indeed, summer 2022 displayed the largest marine heatwave activity of the 1982–2023 period due to an unusually high frequency of long-lasting and intense events, particularly over western Mediterranean.

Taking the June 2022 HW over Iberia as an example, we address the role of dynamical (atmospheric circulation) and thermodynamical (regional soil moisture and western Mediterranean SSTs) drivers in the severity of the event. Flow analogues of the June 2022 heatwave are used to reconstruct the expected temperatures under different combinations of these thermodynamical drivers and assess their separate and combined influences on the intensity of the event. Results show a measurable intensification of the heatwave event (of ~1 °C) by both dry land and warm sea conditions. Although these two drivers are significantly correlated, southwestern European HWs are aggravated if dry soils concur with warm SSTs over western Mediterranean. The magnitude of the Mediterranean SST influence could depend on the soil moisture state, being larger for dry than wet conditions, as well as on the atmospheric circulation.

R.M.T., A.R., S.P. and A.T.M. thank Fundação para a Ciência e a Tecnologia (FCT) I.P./MCTES through national funds (PIDDAC) – UIDB/50019/2020 (https://doi.org/10.54499/UIDP/50019/2020) and LA/P/0068/2020 (https://doi.org/10.54499/LA/P/0068/2020). A.R. and R.M.T. thank also FCT (https://doi.org/10.54499/2022.09185.PTDC, (http://doi.org/10.54499/JPIOCEANS/0001/2019). A.R. was supported by FCT through https://doi.org/10.54499/2022.01167.CEECIND/CP1722/CT0006.

How to cite: Trigo, R., Barriopedro, D., Garrido-Pérez, J., Simon, A., Plecha, S., Teles-Machado, A., Russo, A., and García-Herrera, R.: The outstanding European and Mediterranean heatwave activity during summer 2022, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2460, https://doi.org/10.5194/egusphere-egu24-2460, 2024.

Multiple recent record shattering weather events raise questions about the adequacy of climate models to effectively predict and prepare for unprecedented climate impacts on human life, infrastructure, and ecosystems.

While it appears that  the record breaking global mean temperatures of 2023 are still within the models spread, we show that climate models fail to reproduce the observed emergence of several heatwave hotspots on impact relevant temporal and spatial scales. We identify several heatwave hotspots worldwide in which the warming of the tail of the distribution outpaces the mean warming. In these regions we find that the observed trends by far exceed what is projected by the multi-model mean of state-of-the-art model frameworks. In multiple regions, observed trends are outside of the model spread and persist irrespective of resolution or model design: We investigate models from HighResMip project (49 members) and CAM6 and the ECHAM5 SST forced large ensemble (60 members) and find that biases persist throughout the vast number of climate model experiments with different architectures (fully coupled and forced with observed SSTs), resolutions (25 km – 250km), and large numbers of realizations (Kornhuber et al. submitted).

We discuss potential reasons for the models’ shortcomings. Recently we found that models tend to underestimate the surface anomalies of specific deep atmospheric circulation patterns, which were fundamental to some of the most extreme heat waves in the past (Kornhuber et al. Nat. Comms. 2023). Europe is identified as a heatwave hotspot, driven in part by trends in atmosphere dynamics (Rousi, Kornhuber et al. Nat. Comms. 2022), which models tend to understimate (Vautard et al. Nat. Comms 2023). In addition, non-linear interactions between high pressure, soil moisture deficiencies and temperature have been shown to be pivotal for the record shattering Pacific Northwest heatwave, amplifying its magnitude by 40% (Bartusek, Kornhuber, Ting, Nat. Climate Change 2022). Such dependence structures might not be accurately represented in most models. We conclude that while climate models have been useful and on point regarding their global mean temperature response to increased greenhouse gas concentrations, trends in the far-end tails of the temperature distribution are not well captured, missing emerging high-risk hotspots. This highlights the need to better understand and model the drivers of extreme heat and to rapidly mitigate greenhouse gas emissions to avoid further harm from climate surprises.  

How to cite: Kornhuber, K., Bartusek, S., Seager, R., and Ting, M.: Global emergence of regional heatwave hotspots outpaces climate model projections, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3164, https://doi.org/10.5194/egusphere-egu24-3164, 2024.

How to cite: Mathews, J., Czaja, A., Vitart, F., and Roberts, C.: The Impact of Gulf Stream Moisture Flux Suppression on Atmospheric Blocks, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3633, https://doi.org/10.5194/egusphere-egu24-3633, 2024.

Occurrence of cold spells in different North American regions has been related to concurrent wet and windy extremes in Western Europe. This link is driven by an anomalous state of the North Atlantic storm track. Two dynamical pathways have been defined as potential origins of the Pan-Atlantic compound extremes. The first pathway is linked to a Rossby wave train propagating from the Pacific toward the Atlantic, associated with a pronounced Alaskan ridge. The second pathway is characterized by the presence of a high west of Greenland, that favors simultaneously a southward displacement of a trough over eastern USA and an upper-level trough over South western Europe. The aim of this study is to assess the predictability of these two pathways in the ERA5 reanalysis using dynamical systems indicators. These indicators are the local dimension and the persistence of the large-scale atmospheric flow, and can be used as proxies for the predictability of each pathway. We complement this analysis using the ECMWF ensemble reforecasts at different lead times, and computing skill scores for the two pathways.

How to cite: Krouma, M. and Messori, G.: Assessment of the predictability of cold-wet-windy Pan Atlantic compound extremes, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5138, https://doi.org/10.5194/egusphere-egu24-5138, 2024.

Extreme temperature events can cause severe disruptions to society from negative health consequences to infrastructure damage. Early warning systems are a key element in the mitigation of such impacts, with literature highlighting the potential societal benefit of information from sub-seasonal forecasts. We investigate the relationship between the persistence of an atmospheric state and the practical predictability of surface temperatures focusing on medium to extended range time scales. Persistence is assessed objectively leveraging techniques from dynamical systems theory whilst practical predictability is defined in terms of the forecast error in surface temperature. Atmospheric persistence provides potential value for the practical predictability of temperature in some cases with the results varying depending on season and location. Wintertime temperature forecasts at lead times up to three weeks, and cold spell forecasts up to two weeks in lead time are highlighted as cases where persistence appears to show an association with practical predictability.

How to cite: Holmberg, E., Tietsche, S., and Messori, G.: Persistence and the practical predictability of surface temperatures, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5328, https://doi.org/10.5194/egusphere-egu24-5328, 2024.

Recent years were characterized by spatially co-occurring hot, wet or dry years around the globe. Spatially compounding extreme events can strongly amplify societal impacts as economic supply chains are increasingly interlinked, highlighting the increasing importance of advancing our knowledge of the effects of human-induced climate change on such events. We assess the occurrence of spatially compounding hot, wet and dry years under different future warming levels of 1.5°C , 2°C, 3°C and higher levels of global warming. We focus our analysis on the top-producing agricultural regions that have historically provided the global food systems with large quantities of wheat, maize, soybean or rice. The occurrence of spatially compounding events and area affected in future climates is determined using Earth System Model simulations from the 6th Phase of the Coupled Model Intercomparison Project (CMIP6). The simulations project a strong increase in the global land area that is concurrently affected by hot, wet and dry extremes under continued global warming. On regional scales, the world’s breadbasket regions are particularly affected by strong increases in the simultaneous occurrence of hot, wet or dry extremes under continued global warming. The spatial extent of agricultural land potentially threatened by climate extremes will increase drastically if global mean temperatures shift from +1.5 °C to +2.0 °C, and will be further amplified with every tenth of degree of warming. This highlights that ambitious climate action needs to be taken in order to limit global warming if we want to keep the global agricultural land in a safe climatic space.

How to cite: Biess, B., Gudmundsson, L., Windisch, M., and Seneviratne, S. I.: Future changes in spatially compounding hot, wet or dry events and their implications for the world's breadbasket regions, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5796, https://doi.org/10.5194/egusphere-egu24-5796, 2024.

We investigate statistics of extreme 10 m winds over the midlatitude Atlantic, Pacific and Southern Ocean basins, regions associated with the extratropical storm tracks. We compute  statistics of 10 m wind speed using  reanalysis (ERA5) and satellite scatterometer datasets (NOAA NCEI Blended Seawinds), as well as the outputs from CMIP6 climate models. To select the regions with climatologically strong winds, we study the 10 m wind speeds over the oceans only in the regions where the local 98th percentiles exceed 20 ms ^-1. 

Annually, the median of 10 m wind speed distribution is the highest in the Southern Ocean, while the extreme winds (starting from 90th percentile) are higher over the oceans in the Northern Hemisphere. The hemispheric differences in the extreme winds are greater and more evident during the respective winter seasons, potentially indicating  differences in the dynamics of extreme winter storms. These findings are consistent over all data products analyzed. Additionally, tails of distributions of winds at 850 hPa in the basins during the winter calculated from reanalysis and observations mirror the patterns observed in 10 m wind distributions, pointing to the influence of large-scale processes in creating stronger extreme winds over the Northern Hemisphere.

How to cite: Stanković, A. and Caballero, R.: Statistics of pan-oceanic extreme near-surface winds, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6492, https://doi.org/10.5194/egusphere-egu24-6492, 2024.

European winter windstorms can pose significant risks for the safety and lives of people living in their paths as well as to infrastructure and the natural environment. Several storms in the recent past have caused substantial damages, and risks from extreme winter windstorms may increase with climate change. Characterizing the risks and potential losses from such storms and assessing our ability to predict storm economic losses are therefore an utmost priority. To that end, we have developed a new database of extreme European winter windstorm footprints for the extended winter season (ONDJFM) for the period 1995-2015, and have made it publicly available to both the scientific community and industry. In contrast to previously compiled databases, our database includes storm footprints derived from four different data sets and not from a single source: the ERA5 reanalysis, the COSMO-REA6 reanalysis for Europe, simulation output from a regional climate model driven by ERA5 on the EURO-CORDEX domain, and simulation output from the regional climate model COSMO-CLM on an enlarged Germany domain. We included both the footprints themselves, expressed as the relative daily maximum wind gusts associated with a storm event, and the absolute daily maximum wind gusts associated with that footprint. We derived and included the storm footprints associated with the 50 most extreme storms, or Top50 storms, identified within each of the four input data sets. We applied a consistent methodology for identifying storm footprints and assessing their severity across input data sets that does not require downscaling or adjustment with the assistance of an atmospheric or statistical model. This provides for greater comparability among the footprints derived from the different input sources. Lastly, we derived the Top50 storms from each input on its native horizontal resolution, allowing us to characterize the impact that horizontal resolution can have on footprint identification and severity assessment. Our database thus allows for assessment of extreme storms and their impacts from several perspectives, particularly the impacts from use of wind gust data derived on different horizontal resolutions. This complements the existing extreme European winter storm databases, and facilitates scientific research on extreme storms and industry catastrophe modelling assessments.

How to cite: Flynn, C., Mömken, J., Pinto, J., and Messori, G.: Development of a New Database of Extreme European Winter Windstorms Derived from Multiple Data Sets, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6511, https://doi.org/10.5194/egusphere-egu24-6511, 2024.

European windstorms are among most important natural hazards for insurance companies. Quantifying with the impact of windstorms becomes more challenging due to seasonal loss clustering, characterized by numerous intense windstorms in a season, leading to exceptionally high seasonal losses. Climate change introduces another level of uncertainty regarding potential losses from European windstorm events.

The EURO-CORDEX dataset is designed to enhance the representation of regional and local weather conditions consists of a set of high-resolution climate simulations at 12.5 km resolution.  In this context, this will allow the assessment of the impact of windstorms for recent and future climates in a finer resolution. To achieve this, we use daily maximum surface wind gusts of 20 global-to-regional climate model chains from EURO-CORDEX (EUR-11 domain). The investigation focuses on the extended winter season (ONDJFM) between the historical period (1976-2005) and future projections under global warming level (GWL) scenarios of +2°C and +3°C, following the Representative Concentration Pathway 8.5 (2006-2100).

The evaluation of windstorm impact is carried out using the Loss Index (LI) method, focusing on the country level. For the historical period, a substantial bias is observed in the 98th percentile of daily maximum wind gusts between EURO-CORDEX and ERA5. This bias is corrected through empirical quantile mapping, resulting in corrected models that show reduced biases in wind gust extremes while maintaining consistency with the climate change signal.

Under the +2°C and +3°C GWLs, the majority of models indicates a reduction in the magnitude and frequency of extreme windstorms over Western Europe and the Iberian Peninsula, leading to decreased European windstorm loss, while an increase over Eastern Europe is expected, contributing to higher loss. In the majority of countries, the occurrence of seasonal loss clustering is expected to decrease under GWL conditions compared to the current climate.

Our study provides valuable insights for insurance companies and policymakers to deal with the uncertainty of the loss of windstorm under future climate conditions.

How to cite: Alifdini, I., Moemken, J., Ramos, A. M., Stankovic, A., Caballero, R., and Pinto, J. G.: Future Changes in European windstorm loss and seasonal loss clustering in the EURO-CORDEX dataset, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7539, https://doi.org/10.5194/egusphere-egu24-7539, 2024.

Ongoing global warming has major implications on the food security and the agricultural sector. Extreme weather such as precipitation extremes, heatwaves, frost and drought can cause severe reductions in crop production. Furthermore, these events can be strongly amplifyed when considered as compound events. In this study, we inquire into the temporally combination of climate extremes leading to substantial drops in winter wheat yield (in the following called “yield shock “) throughout different countries in Europe. Winter wheat is one of the most important crops in Europe in terms of production, and is also acknowledged a major crop globally. We use the Global Data of Agricultural Yield, a satellite-reported yield hybrid dataset of major crops. We categorise the yield as “yield shock” and “no yield shock” in order to reduce the uncertainty and to mainly focus on historical yield plunges. We consider and test different climate indicators (like the number of warm days or cumulative precipitation) that represent weather extreme events at a subseasonal scale. Moreover, we employ a Random Forest to capture any possible nonlinear relation. Our study illustrates the probability of winter wheat yield shock under the occurrence and co-occurrence of subseasonal weather extremes, and the nuances throughout countries with different climatic patterns.

How to cite: Stainoh, F., Biess, B., Gudmundsson, L., Moemken, J., Seneviratne, S. I., and Pinto, J. G.: Assessing the climate drivers leading to winter wheat yield shock in Europe, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9093, https://doi.org/10.5194/egusphere-egu24-9093, 2024.

Machine Learning (ML) encompasses various techniques and algorithms that have proven highly effective in addressing complex climate science tasks. In particular, using ML to detect and forecast extreme events has gained much attention recently. Considering the vast volume of spatial and temporal data available, the employment of data-driven methodologies becomes indispensable for effectively uncovering potential drivers of these events. This study arises with the ambition of proposing a comprehensive and general framework that provides insights about interactions between heatwaves and potential physical drivers across multiple spatio-temporal scales. A novel feature-selection methodology is presented to advance the detection of heatwaves. It is based on a two-step procedure. In the first stage, a non-supervised task is developed for spatially clustering the different variables. Thus, there is a reduced initial pool of driver candidates. In the second stage, a wrapper methodology is applied to determine which time periods are representative for each of the clusters in the occurrence of heatwaves. This algorithm discerns which clusters are selected as drivers and which are discarded, along with the time period, relative to the heatwave occurrence, during which each cluster should be investigated. Thus, the feature selection is developed based on the spatio-temporal distribution of the different variable clusters.

Experiments have been developed for detecting heatwaves in the Lake Como region, a region of key agricultural activity in Northern Italy. A  wide range of ocean and atmospheric variables taken from ERA5 are used (e.g. sea ice concentration, precipitation). The framework allows the identification of the time lag for each variable from short-term to seasonal time scales (up to 180 days). Results have spotted drivers on the subseasonal to seasonal timescale. Important drivers have been identified for short-term periods (less than one week). Local variables are shown to be of much significance for these periods. The method allows for identifying the strong influence of local variables whilst identifying correlations between the heatwave occurrence and different variables and spatial locations. The importance stages of the variable candidates can be established by running the model while removing the most critical variables.

How to cite: Peláez-Rodríguez, C., Pérez-Aracil, J., McAdam, R., Squintu, A., Scoccimarro, E., and Salcedo-Sanz, S.: A Spatio-Temporal Optimization- Based Feature Selection Framework for detecting drivers of heatwaves, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10153, https://doi.org/10.5194/egusphere-egu24-10153, 2024.

Extreme climate events have major impacts on human societies and ecosystems. The most detrimental events are often extremely rare, with return time of centuries or even millennia. Studying these events in the context of climate change is crucial to help adaptation efforts globally. Yet the study of these extremely rare events is extremely challenging due to the lack of data. Indeed, such events have likely not been observed in the instrumental period. An alternative is to use a global climate model to simulate these extremely rare events. However, this comes at a huge computational cost : gathering good statistics on centennial events would require to run a few thousand years of simulation. 

Rare event algorithms have recently been introduced in the field of climate science to tackle this difficulty [1]. By concentrating the computational effort on the trajectories most susceptible to lead to the extreme event of interest, they allow for the sampling of extremely rare events at a much lower computational cost than standard simulations. 

In this study, we run a rare event algorithm to sample extreme heatwave seasons in a heatwave hotspot of South Asia, using the intermediate complexity model Plasim. We compare the outcome of the algorithm against an extremely long – 8000 years – control run. This comparison allows us to demonstrate that the algorithm not only estimates return times with high precision (as shown in previous work), but also exhibits high precision in the estimation of composite statistics: composite maps conditioned on centennial heatwave seasons estimated from the algorithm, are in very good agreement with the ones from the 8000-year long control simulation. Our results suggest that extreme heatwave seasons in the studied region are associated with a quasi-stationary atmospheric wave-pattern stretching from the North Atlantic towards South Asia. 

We also show that the algorithm correctly estimates the intensity-duration-frequency statistics of subseasonal heatwaves occuring within centennial heatwave seasons. Thus rare event algorithms could, for instance, be combined with seasonal forecasts to provide information regarding expected number of heatwave days and the distribution of the duration and intensity in an extreme heatwave season, which could be useful for adaptation planning. 

References

[1] F. Ragone, J. Wouters, and F. Bouchet, “Computation of extreme heat waves in climate models using a large deviation algorithm,” Proc Natl Acad Sci USA, vol. 115, pp. 24–29, Jan. 2018. 

How to cite: Le Priol, C., Monteiro, J. M., and Bouchet, F.: Study of extreme heatwave seasons in South Asia using rare event simulations, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10586, https://doi.org/10.5194/egusphere-egu24-10586, 2024.

How to cite: Noyelle, R., Zhang, Y., Yiou, P., and Faranda, D.: Maximal reachable temperatures for Western Europe in current climate, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11046, https://doi.org/10.5194/egusphere-egu24-11046, 2024.

Events of extreme precipitation pose a hazard to many parts of Europe but are typically not well represented in climate models. Here, we evaluate daily extreme precipitation over Europe during 1982-2019 in observations (GPCC), reanalysis (ERA-5) and a set of atmosphere-only simulations at low- (100km), medium- (50km) and high- (25km) resolution with OpenIFS (Version43R3). We find that both model simulations and reanalysis underestimate the rates of extreme precipitation compared to observations. The biases are largest for the lowest resolution (100 km) and decrease with increasing horizontal resolution (50 and 25 km) in all seasons. The sensitivity to horizontal resolution is particularly high in mountain regions, likely linked to the sensitivity of vertical velocity to the representation of topography. The sensitivity of precipitation extremes to model resolution increases dramatically with increasing percentiles, which modest biases at the 70^th percentile and large biases at 99^th percentile.  We also find that extreme precipitation mostly consists of large-scale precipitation (~80%) in winter, while in summer it is mostly large-scale precipitation in Northern Europe (~70%) and convective precipitation in Southern Europe (~70%). Compared to ERA5, the model simulations produce higher large-scale precipitation extremes in winter, but weaker in summer. The discrepancy between OpenIFS simulations and ERA-5 decreases with increasing horizontal resolutions. We also examine the model time step’s effect on extreme precipitation. The results show that the convective contribution to extreme precipitation is more sensitive to the model time step than horizontal resolution. This is likely due to the sensitivity of convective activity to model time step. On the other hand, the large-scale contribution to extreme precipitation is more sensitive to horizontal resolution than model time step, which may be due to sharper fronts and steeper topography at higher resolution. In general, the lowest-resolution and longest time step has overall higher biases than the highest-resolution and shortest time step.

How to cite: Liu, Y., Kjellsson, J., Savita, A., Lübbecke, J., and Park, W.: Impact of horizontal resolution and model time step on European precipitation extremes in the OpenIFS atmosphere model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11183, https://doi.org/10.5194/egusphere-egu24-11183, 2024.

Heatwaves heavily affect European public health, society and economy. A full understanding of the drivers behind the occurrence and intensity of heatwaves (HW) is one of the priorities of H2020 CLimate INTelligence (CLINT) project. Particular attention is given to the detection and attribution of HWs in future climate projections. However, it is important to assess the capability of climate models to thoroughly describe relationships between the drivers and the occurrence and intensity of HWs. For this reason, a feature selection framework, based on the Coral Reef Optimization (Salcedo-Sanz et al., 2014) has been developed. This has been applied to ERA5 summer data, using as a target the Lake Como HW occurrence and, as candidate predictors, time series of weather variables calculated on clustered areas on European and global scales. The same algorithm has been applied to historical climate simulations included in CMIP6. The comparison of the results of these two steps has first focused on the similarities in maximum temperature and HW trends in the target region of Lake Como. Then, the selected drivers in each historical climate simulation have been evaluated, using ERA5 results as benchmark. Thanks to this, the models that better resemble the statistical properties and teleconnections described by the reanalysis have been identified. This set of models will be considered for the detection and attribution of heatwaves in future climate projections under different emission scenarios. In this upcoming phase the goal will be to analyse changes in the relationship between the drivers and HW occurrence and intensity, giving an insight about possible future evolutions in heatwaves frequency and magnitude.

How to cite: Squintu, A. A., McAdam, R., Peláez Rodríguez, C., Pérez Aracil, J., Alvarez Castro, C., and Scoccimarro, E.: Drivers of heatwaves in CMIP6 models: an evaluation on historical simulations, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11807, https://doi.org/10.5194/egusphere-egu24-11807, 2024.

Theoretical and numerical studies have shown that transient atmospheric motions leading to weather extremes can be classified through the instantaneous dimension and stability of a state of a dynamical system [Faranda et al., Sci. Rep., 2017]. The asymptotic values of these quantities can be computed theoretically only for specific systems, while their numerical counterpart for climate observables provides information on the rarity, predictability, and persistence of specific states. In this work, we present a first attempt to relate the presence of extreme events with the elements that make up a templex of the system under study, both in the deterministic [Charó et al., Chaos, 2022] and stochastic frameworks [Charó et al., Chaos, 2023]. The templex provides the key characteristics of the topological structure underlying a dynamical system. This work will present results for the classical, deterministic Lorenz [JAS, 1963] attractor and for the Lorenz Random Attractor, dubbed LORA [Ghil & Sciamarella, NPG, 2023].

How to cite: Charó, G. D., Faranda, D., Ghil, M., and Sciamarella, D.: Extreme events in a templex, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11997, https://doi.org/10.5194/egusphere-egu24-11997, 2024.

Windstorms are extreme midlatitude cyclones and one of the major natural hazards that cause damage and losses in Europe. While the processes involved in their genesis and intensification are generally well understood, there are still considerable uncertainties in the estimation of associated impacts like widespread wind damage and flooding. The compounding characteristics of the events further enhances the complexity of this task. This is even more true for the impact forecasting of windstorms on weather and sub-seasonal time scales. Additionally, there are large uncertainties on how windstorms and their impacts will change in a warmer climate, particularly regarding the role played diabatic processes in a warmer atmosphere. This study presents examples of recent developments regarding windstorms and discusses some new avenues for interdisciplinary research towards bridging the gap between fundamental research and practical applications.

How to cite: Pinto, J. G.: European Windstorms: bridging the gap between fundamental research and practical applications, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12471, https://doi.org/10.5194/egusphere-egu24-12471, 2024.

The presence of anomalously cold air close to the surface is a prerequisite for the occurrence of cold weather extremes, such as cold spells. However, the process of cold air generation features a substantial variability in space and time, modulated by the rate of energy loss to space by infrared radiation.

Such a variability is investigated using a Lagrangian approach, identifying trajectories that experience rapid non-adiabatic cooling over Eurasia. This approach allows to identify source regions of cold air, and the meteorological conditions that particularly favor its generation and accumulation – which often precedes the most extreme cold spells.

The unraveling of this connection allows to interpret the intra-seasonal and the inter-annual variability in the occurrence of cold extremes, and to gain a mechanistic understanding of how anthropogenic global warming will modify them.

How to cite: Riboldi, J.: Quantifying the generation of cold air during boreal winter and its relevance for cold weather extremes, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-13211, https://doi.org/10.5194/egusphere-egu24-13211, 2024.

As we attempt to infer a system's response to external driving from measurements, random errors in the measurement of the drivers can lead us to mistakenly infer a non-linear response. In particular, we are likely to underestimate the system's response during extreme and rare driving conditions due to uncertainty in the drivers. We demonstrate this phenomenon for extreme space weather and its impact on Earth's magnetosphere, where due to random errors in the measurements of solar wind drivers, there is a non-linear bias in the magnetosphere's response. We propose that the underlying statistical effect (related to the more well-known regression to the mean effect) is generalizable to the other fields that study different systems' responses to driving, like extreme climate studies. 

How to cite: Sivadas, N. and Sibeck, D.: Underestimating extremes due to random uncertainty, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-13645, https://doi.org/10.5194/egusphere-egu24-13645, 2024.

During the last years, the statistical analysis of compound extremes has gained increasing interest among the scientific community due to the multiple threats presented by such events to society, economy, and the environment. In many situations, the statistics of compound extremes is based on bivariate extreme value theory and measures provided by this framework. Such choice of statistical methodology may however not properly address two relevant aspects: the non-zero duration of such events (which can be rather persistent, e.g. in the case of droughts or heatwaves, which heavily violates the independence assumption of classical extreme value theory) and the fact that not all events of practical revelance can be described as cases falling into the tails of the distribution of some observable of interest.

A general framework addressing the non-extremeness aspect is provided by event coincidence analysis (ECA), which quantifies the empirical frequency of co-occurring events of arbitrary types and allows ist comparison with the values for certain random null models like independent Poisson processes with prescribed event rates. While classical ECA is based on temporal point processes and hence may be criticized for not capturing the statistical characteristics of persistent events very well, I will present a new methodological variant, interval coverage analysis (InCA), as an alternative for specifically addressing co-occurrences of persistent events. In the limit of vanishing event durations, the new interval coverage rates of InCA are identical to the event coincidence rates provided by ECA. By allowing for mutual time shifts between the different types of events under study as well as a temporal tolerance regarding their respective timing, fixed and even distributed time lags can be taken into consideration.

This presentation introduces and compares the basic methodological concepts behind both ECA and InCA (including their extension to studying multivariate and conditional dependency), and demonstrates an example of their respective application in geoscientific contexts. Specifically, the spatial patterns of dependency between the timing of heatwaves and large-scale circulation anomalies of the atmospheric jet stream in the Northern hemisphere is studied. The corresponding analysis reveals specific regions with elevated likelihood of heatwaves along with the emergence of a split (double-banded) jet stream, while the emergence of heatwaves is suppressed at the same time in other regions. The obtained results may thus guide further targeted research regarding the specific mechanisms leading to this regional differentiation in heatwave frequency.

How to cite: Donner, R. V.: Studying statistical dependency among short and persistent events – recent developments and application to mid-latitude circulation anomalies associated with heatwaves, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-14476, https://doi.org/10.5194/egusphere-egu24-14476, 2024.

Over the last decade, researchers have devoted considerable efforts to exploring the feasibility of attaining convection-permitting scales in regional climate model projections. The primary objective has been to comprehend the effects of global warming on climate extremes. Initially conducted as isolated model experiments, these investigations have evolved into coordinated ensemble experiments operating at convection-permitting scales across diverse continents. The implementation of such coordinated ensembles has provided a platform for evaluating model reliability at high resolutions and conducting signal-to-noise analyses on identified climate change signals.

Nevertheless, the constraints imposed by limited computational resources have confined these experiments to smaller domains compared to the conventional continental scale employed in dynamical downscaling, as seen in initiatives like the CORDEX community and time slice mode.

Despite these inherent limitations, these experiments have successfully showcased the models' capacity to simulate present-day climate conditions. Notably, improvements in various statistical metrics at sub-daily scales have been observed in contrast to parametrized models. Furthermore, the ensemble approach has contributed to reducing uncertainties in assessing both present-day climate and future projections, particularly in terms of frequency, intensity, and extreme precipitation at the hourly time scale.

The explicit representation of convection has additionally enabled the study of convective storm system evolution, allowing for an assessment of large-scale feature changes and related physical mechanism driving observed extreme precipitation variations.

Preliminary attempts have also been carried on for building convection permitting climate emulator to reduce the computational HPC demand required by the dynamical downscaling models.

These collective findings underscore the necessity of advancing to the next phase of coordination, involving the establishment of multiple coordinated platforms spanning different continents. These platforms will serve as collaborative spaces for discussing model enhancements, aiming to refine existing models by incorporating a more precise representation of Earth system components and defining domains that maximize the number of models capable of generating convection-permitting climate projections.

How to cite: Coppola, E.: Understanding global warming impact on climate extremes by mean of Coordinated Ensemble Experiments of Convection-Permitting Climate Projections, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-16130, https://doi.org/10.5194/egusphere-egu24-16130, 2024.

Heatwaves rank among the most devastating extreme events over Europe, particularly in terms of mortality and agricultural damage. Heatwaves are often defined as exceedances of thresholds based on daily maximum temperatures, thus considering only daytime effects. However, exceedances of daily minimum temperatures, which often occur at night, are of similar importance, as they can cause additional stress to the human health due to shorter recovery times. For both daytime and nighttime heatwaves, knowledge of the underlying mechanisms is crucial for the successful prediction of the events; however, these mechanisms are not yet fully understood.

Although heatwaves can occur quite locally, the task of heatwave prediction can be simplified by representing heatwaves as recurring large-scale patterns. The aim of the presented work is to identify these dominant coherent spatial patterns over Europe using the SANDRA (Simulated Annealing and Diversified Randomization) clustering method for both daytime and nighttime heatwaves. In a second step, relevant atmospheric, oceanic and land variables are investigated for their physical connection to each of the European heatwave clusters with different time lags.

The results obtained from reanalysis data covering the recent past are compared with those obtained from a long coupled global climate model simulation of the last 2000 years, performed with the MPI-ESM model. The long model simulation is further used to place the recent record-breaking European heatwave of summer 2022 in a historical context.

How to cite: Hansen, F., Feser, F., and Zorita, E.: Day- and nighttime heatwave clusters over Europe and their physical drivers, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-16462, https://doi.org/10.5194/egusphere-egu24-16462, 2024.

Mid-tropospheric deep depressions in summer over the North Atlantic are shown to have strongly increased in the eastern and strongly decreased in the western North Atlantic region.  This evolution is linked to a change in baroclinicity in the west of the North Atlantic ocean and over the North American coast, likely due to the increased surface temperature there. Deep depressions in the Eastern North Atlantic are linked to a temperature pattern typical of extreme heat events in the region. The same analysis is applied to a sample of CMIP6 model outputs, and no such trends are found.  This study suggests a link between the observed increase of summer extreme heat events in the region and the increase of the number of Atlantic depressions. The failure of CMIP6 models to reproduce these events can consequently also reside in an incorrect reproduction of this specific feature of midlatitude atmospheric dynamics.

How to cite: D'Andrea, F.: Summer Deep Depressions Increase Over the North Atlantic, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-16848, https://doi.org/10.5194/egusphere-egu24-16848, 2024.

Intraseasonal variability of extreme rainfall events (EREs) during the South Asian Summer Monsoon season is modulated by the Boreal Summer Intraseasonal Oscillation (BSISO), a convective system of organised heavy rainfall that moves periodically from the Indian Ocean to the Western Pacific over the subcontinent. The BSISO, in turn, is typically characterised using indices obtained from the leading components of an Empirical Orthogonal Analysis (EOF) of outgoing longwave radiation (OLR) and lower and upper troposhere wind data from the region [1] ). A primary motivation for using OLR and wind data is that the EOF-analysis is not well suited for heavy-tailed data such as precipitation. 

Here, we propose to estimate climate indices directly from EREs in the South Asian Summer Monsoon by applying the Hidden Climate Index (HCI) - framework introduced by Renard et al. (2022) [2]. The method is designed to work with binary, event-like data and utilizes a Bayesian hierarchical model incorporating spatial Gaussian process priors, to capture spatial and temporal interdependencies by sampling from a Bernoulli distribution.

Using the HCI framework, we estimate latent variables that underlie the ERE dynamics in the South Asian Monsoon domain, and show that these are related to large-scale modes of climate variability., We demonstrate that the ERE-based HCIs correlated well to the BSISO and in addition, we find relationships between the observed large-scale spatial ERE patterns  to the El Nino Southern Oscillation and the Silk Road Pattern.

[1] Kikuchi, K., Wang, B. & Kajikawa, Y. Bimodal (2012). Representation of the tropical intraseasonal oscillation. Clim. Dyn. 38, 1989–2000.

[2] Renard, B., Thyer, M., McInerney, D., Kavetski, D., Leonard, M., & Westra, S. (2022). A Hidden Climate Indices Modeling Framework for Multivariable Space-Time Data. Water Resources Research, 58

How to cite: Früh, D., Strnad, F., and Goswami, B.: Climate Indices of Extreme Rainfall in the South Asian Monsoon Domain using a Bayesian Hierarchical Model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17701, https://doi.org/10.5194/egusphere-egu24-17701, 2024.

Recent work has provided robust evidence for the systematic co-occurrence of wintertime cold spells in North America and wet and windy extremes in Europe, which we term compound pan-Atlantic extremes. Both cold spells and wet and windy extremes are individually highly impactful, and their concurrence further amplifies their effects for actors with international exposure who are vulnerable to correlated losses. This study aims to investigate further the atmospheric processes associated with compound pan-Atlantic cold, wet and windy extremes and how these processes are represented in CMIP6 models.

On aggregate, cold spells in different parts of North America statistically co-occur with wind and precipitation extremes in specific European regions. However, North American cold spells can arise from multiple dynamical pathways, altering the location and timing of the associated European extremes for individual cold spells. Here, we use ERA5 reanalysis data (1940-2014) to identify North American wintertime cold spells in three different regions and relate the occurrence of European extremes to Pacific and Atlantic weather regimes. We further analyze the various pathways of North American cold spells by evaluating the relative contribution of planetary (k=1-3) and synoptic (k=4-8) Rossby waves to the resultant weather regimes. The evaluation is performed by partitioning the Rossby wave circulation into different zonal wavenumber ranges using the MODES software, based on the normal-mode function decomposition. This methodology has previously been employed to identify changes in the midlatitude circulation at multiple scales during Eurasian heatwaves, though it is novel in its application to cold spells. Here, we discuss how the wavenumber ranges differ across cold spells and from the climatological state before and during the cold spells. We next compare the CMIP6 historical simulation model data (1940-2014) with the ERA5 results. First, we review the ability of the models to replicate the spatial and temporal pattern of pan-Atlantic extremes for the three cold spell regions. Second, we discuss the performance of the models in capturing the weather regime frequencies and the planetary and synoptic Rossby wave contributions to the North American cold spell pathways.
The results of this study contribute to the evaluation of the model fidelity in reproducing pan-Atlantic compound extremes and the associated circulation, with direct implications for the assessment of climate projections.

How to cite: Leeding, R., Strigunova, I., and Messori, G.: Dynamics of pan-Atlantic winter compound extremes in ERA5 and CMIP6 models, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18308, https://doi.org/10.5194/egusphere-egu24-18308, 2024.

Low-frequency variability (LFV) encompasses atmospheric and climate processes on time scales from a few weeks to decades.​ This includes atmospheric blockings, heat waves, cold spells, and at longer time scales long-term oscillations like the MJO, the NAO, ENSO….. Better understanding of LFV, could contribute to improved long term forecasts​. 

In the results described in Xavier et al, 2023, who used a reduced order atmosphere-land model (Demaeyer et al, 2020), weather patterns that involve atmospheric blocking to the west of a given topographical feature tend to have reduced predictability and show instability when contrasted with blocking occurrences situated to the east of such topographical elements. This finding aligns with actual meteorological occurrences, such as the persistence of North Pacific blocking patterns (Breeden et al., 2020; Kim and Kim, 2019). The shape and characteristics of the identified blocking events closely resemble North Pacific blocks, where a high-pressure system exists either on the western or eastern side of the underlying topography. In the physical world, these positions correspond to Asian and American continents on either side of the Pacific. Despite quasi-geostrophic models being overly simplified, using such reduced order models in this study allowed us to undertake such mathematical analysis. Thus allowing for a comparison with the real world…

In the current study, we aim to analyze these predictability differences based on the morphology of the blocking situations in the real-world scenario using the CMIP6 dataset. Blocking situations are identified using two different indices, the anomaly-based blocking index proposed by Sausen et al., 1995 and the local reversal of the meridional flow-based index proposed by Davini et al., 2012. Predictability is quantified using local dimension metrics and analogue studies in the identified blocking events. The findings are discussed from the perspective of the current literature on the predictability of blocking.

References

Xavier, A. K., Demaeyer, J., and Vannitsem, S.: Variability and Predictability of a reduced-order land-atmosphere coupled model, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2023-2257, 2023.

Demaeyer, Jonathan & De Cruz, Lesley & Vannitsem, S.: qgs: A flexible Python framework of reduced-order multiscale climate models. Journal of Open Source Software. 5. 2597. 10.21105/joss.02597, 2020.

Breeden, M. L., Hoover, B. T., Newman, M., and Vimont, D. J.: Optimal North Pacific blocking precursors and their deterministic subseasonal evolution during boreal winter, Monthly Weather Review, 148, 739–761, 2020

Kim, S.-H. and Kim, B.-M.: In search of winter blocking in the western North Pacific Ocean, Geophysical Research Letters, 46, 9271–9280, 2019.

Sausen, R., König, W., & Sielmann, F. (1995). Analysis of blocking events from observations and ECHAM model simulations. Tellus A, 47(4), 421–438.

Davini, P., Cagnazzo, C., Gualdi, S., & Navarra, A. (2012). Bidimensional diagnostics, variability, and trends of Northern Hemisphere blocking. Journal of Climate, 25(19), 6496–6509.

How to cite: K Xavier, A., Hamilton, O., Faranda, D., and Vannitsem, S.: Investigating the Influence of Atmospheric Blocking Morphology on Predictability, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18316, https://doi.org/10.5194/egusphere-egu24-18316, 2024.

The unprecedented changes in Earth's climate are reshaping atmospheric dynamics on a global scale, with profound implications for various sectors, including aviation. One of the critical facets of this transformation lies in the alterations to atmospheric circulation patterns and the concurrent modifications in turbulence characteristics.

We investigate the intricate interplay between climate change, atmospheric circulation patterns, and turbulence modifications over Europe, with a specific focus on their implications for commercial flight operations. Drawing upon ERA5 reanalysis data at 200 hPa pressure level we explore the evolving climatic conditions shaping the European airspace. By jointly looking at the cube-rooted eddy dissipation rate, vorticity, and horizontal divergence we evidence increasing trends of anomalously turbulent conditions over UK and Northern Europe. Otherwise, decreasingly frequency anomalies are associated with light-turbulent conditions over Central and Mediterranean Europe. Overall, increasing trends also correspond to increased severity in terms of turbulence strength levels, with a clear increase in moderate-to-severe turbulence episodes, mainly associated with both converging and diverging atmospheric ridges.

The study highlights the region's vulnerability to significant changes in atmospheric pattern dynamics, emphasizing the potential increase in turbulence-related episodes and severity, impacting aviation safety, fuel efficiency, and passenger comfort. Our analysis aims to provide valuable insights for aviation stakeholders, policymakers, and researchers, contributing to the development of adaptive strategies and operational guidelines.

How to cite: Alberti, T.: Atmospheric circulation changes and turbulence modifications over Europe: implications for commercial flight operations in a changing climate, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18608, https://doi.org/10.5194/egusphere-egu24-18608, 2024.

Filomena was an extratropical cyclone in early January 2021 that was most notable for bringing unusually heavy snowfall to parts of Spain, with Madrid recording its heaviest snowfall in over a century, and with Portugal being hit less severely. Filomena caused severe winds, heavy rainfall and snowfall in different parts of the country but also occurring at the same time. The atmospheric pattern during the event of Filomena was a rare one characterized by a low intrinsic predictability. Despite the number of studies focusing on the detection and characterization of extreme events in Western Europe, our knowledge regarding the predictability of such occurrences remains limited. By studying the intrinsic predictability of an extreme event, we can know the capacity we have to anticipate it, thus being able to take action for the minimization of its impacts by using early warning systems.

In this work, we show first an overview of the intense low pressure systems (including Filomena) that have recently struck specific cities/regions of the Iberian Peninsula with concurrent events. We assess their predictability and, finally, we perform an attribution study of these events to climate change.

How to cite: Alvarez-Castro, C., Faranda, D., Gallego, D., Peña-Ortiz, C., Torralba, V., and Gualdi, S.: Predictability of the low pressure systems leading concurrent events in the Iberian Peninsula., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18695, https://doi.org/10.5194/egusphere-egu24-18695, 2024.

Heat waves are a growing issue in the current climate, causing damage to human societies and other living beings. As climate warms, heat waves are one of the extreme events that will be exacerbated by the rising average temperatures. Understanding the mechanisms that drive heat waves is hence of vital importance to analyze them and to make predictions to mitigate their impacts. However, the rarity of any extreme event makes it particularly hard to study, especially if one wishes to understand the relation between predictors and probability, for instance using machine learning techniques, which are notoriously data-hungry. A possible solution is to use climate model simulations, but they introduce biases and can be prohibitively expensive to run for appropriate dataset lengths.

In this work, we introduce a simple and powerful statistical model, that is able to skillfully predict rare heat waves in a regime of lack of data, as it is the case for the ERA5 reanalysis dataset.

We focus on two-week heat waves over France, and we notice that composite maps of very extreme events are similar to the ones of much less extreme ones. This holds true for an increasing hierarchy of model complexity, including ERA5. We can thus analyze very extreme events by looking at less rare ones, having the advantage of increasing the available statistics. This effect can be explained assuming that the set of predictors and the heat wave amplitude are jointly gaussian. The prediction task can be thus rephrased into estimating the conditional probability of an extreme heat wave happening conditioned on the state of the set of predictors. This quantity, called committor function, is generally hard to estimate given the high dimensionality of the variables involved. Our assumption performs a dimensionality reduction, where an estimate is obtained through an optimal linear projection. The projection map will suggest the most relevant predictors.

We first demonstrate that this Gaussian approximation performs well on a 8000 years run of a model of intermediate complexity, PlaSim, for both analysis and prediction tasks when 500 hPa geopotential height, temperature and soil moisture are used as input fields. When we compare with a machine learning estimation of the committor, which is possible here thanks to the huge amount of available data, our Gaussian model is less skillful, but not by much. We then apply the Gaussian assumption to the 500hPa geopotential height field of the ERA5 dataset, outperforming the neural networks and extending the predictability horizon by several days.

The prediction method we propose is simple and effective. It opens possibilities to achieve skilfull predictions that outperform all known approaches, for instance machine learning, notably when data is scarce. Our method also serves as a better baseline than the climatology to benchmark more complex approaches. Since our statistical model is also interpretable, this framework has potential to go beyond prediction skill only and, thanks to the optimal projection map, foster the study of fast and slow drivers, and the effect climate change has on them.

How to cite: Mascolo, V., Lovo, A., Herbert, C., and Bouchet, F.: A Gaussian framework for optimal prediction of extreme heat waves, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18866, https://doi.org/10.5194/egusphere-egu24-18866, 2024.

Recent studies have shown two-way interactions between aloft large-scale structures in the atmosphere
and local features such as surface temperature, wind, and land use. This requires the use of high-resolution land use schemes and convection-permitting models (CPM) for large eddy simulations (LES).  Weather
Research and Forecasting model (WRF) is being increasingly used with resolutions that allow convection
to be fully simulated, and efforts (such as CORDEX FPS URB RC) have been made to introduce more
precise land use schemes to model the impacts of urban zones on temperature and wind statistics. In
this study, we focus on the 2003 summer heat wave and compute temperature and wind statistics from
surface to upper tropospheric pressure levels, ranging from microscales (~50m) to mesoscales (~500km) in
North-western France. The emphasis is put on extreme values of temperature by computing its Probability
Density Function (PDF) over the domain and across different spatial scales. Results pertain to second,
third, and fourth-order moments of temperature and wind reflecting variance, direction of across-scale
interactions, and extreme events' occurrence probability, respectively. Finally, we correlate mean large-scale
temperature gradients with those extreme events. This study provides new insights into the complex and
continuous across scales two-way interactions between local features and large-scale climate.

How to cite: Danaila, L., Blervacq, C., Sayeed, K., Fossa, M., Massei, N., and Chun, K.: Temperature and wind statistics with convection-permitting model WRF in the context of heat waves and urban heat island events, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-20380, https://doi.org/10.5194/egusphere-egu24-20380, 2024.

Earth-system Models of Intermediate Complexity (EMICs) are climate models featuring a simplified representation of climate processes and a much lower computation cost. This makes them particularly suitable for exploring phenomena with a large ensemble simulation approach. Here we use the coupled atmosphere-ocean PlaSIM-LSG EMIC to study the possibility of Atlantic Meridional Overturning Circulation (AMOC) spontaneous collapses and how this is altered in the presence of external anthropogenic forcing. Understanding the stability of the AMOC and its response to anthropogenic forcing is of key importance for advancing climate science. The idea of a “safe-operating space” has been proposed in order to define a threshold on anthropogenic forcing within which the AMOC does not lose stability. This requires understanding the combined action of CO2-driven and noise-induced processes in climate tipping events

 First, we address the occurrence of noise-induced AMOC collapses, i.e. spontaneous abrupt weakening  events induced by chaotic internal climate variability in absence of any external forcing. We address the problem of finding these extreme events via the application of a Rare Event Algorithm, which - via a selective cloning of the most interesting model trajectories -  allows a faster exploration of the model phase space in the direction of an AMOC decrease. The algorithm is applied to a PlaSIM-LSG ensemble simulation run at T21 spectral resolution in the atmosphere, and 3.5 degrees in the ocean, with fixed pre-industrial conditions. A number of collapse events, unseen in the pre-industrial control run, are sampled by the algorithm. Looking at the mechanisms causing the AMOC spontaneous collapse, we find that zonal wind stress over the North Atlantic is the main driver of the initial AMOC slowdown, while the suppression of surface convection in the Labrador sea is the likely cause of the subsequent AMOC collapse. Then, we investigate the influence of increasing CO2 levels on the frequency of these spontaneous AMOC collapses. We show that a higher CO2 not only leads to the well-known weakening of the AMOC mean state, but it also increases the possibility of incurring in abrupt noise-induced transitions. The employment of EMICs, combined with the proposed approach, samples a large number of rare phenomena. This procedure allows us to explore statistical properties that are not accessible with a deterministic approach in state-of-the-art high resolution models.

How to cite: Cini, M., Zappa, G., Ragone, F., and Corti, S.: Exploring Noise-induced and CO2-driven AMOC collapses in the PlaSIM-LSG climate model with a Rare Event Algorithm., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1036, https://doi.org/10.5194/egusphere-egu24-1036, 2024.

The search of early-warning signs able to predict the approach of a parameter to a deterministic bifurcation threshold is relevant in climate as it aims to enable a proper prediction of qualitative changes in the studied models. The observation of such objects in SPDEs (stochastic partial differential equations) permits the consideration of space variables and the ensuing heterogeneity in the behaviour of their solutions.

The presence of Gaussian noise on the boundary of the studied space is used in order to build the signals, whose properties are discussed thoroughly. An example in the form of application of such tools on a climate model is presented and justified. The utility and appropriate use of the results on a more applied perspective are shown.

How to cite: Bernuzzi, P., Kuehn, C., and Dijkstra, H.: Early-Warning Signs for SPDEs under Boundary Noise, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2167, https://doi.org/10.5194/egusphere-egu24-2167, 2024.

Aeolus 2.0 is an open-source numerical atmosphere model with intermediate complexity designed to capture the dynamics of the atmosphere, especially extreme weather and climate events. The model's dynamical core is built on a novel multi-layer pseudo-spectral moist-convective Thermal Rotating Shallow Water (mcTRSW) model, and it utilizes the Dedalus algorithm, renowned for its efficient handling of spin-weighted spherical harmonics in solving pseudo-spectral problems. Aeolus 2.0 comprehensively characterizes the temporal and spatial evolution of key atmospheric variables, including vertically integrated potential temperature, thickness, water vapor, precipitation, and the influence of bottom topography, radiative transfer, and insolation. It provides a versatile platform with resolutions ranging from smooth to coarse, enabling the exploration of a wide spectrum of dynamic phenomena with varying levels of detail and precision.

The model has been utilized to investigate the adjustment of large-scale localized buoyancy anomalies in mid-latitude and equatorial regions, along with the nonlinear evolution of key variables in both adiabatic and moist-convective environments. Our findings highlight the triggering mechanisms of phenomena such as the Madden-Julian Oscillation (MJO) and the circulation patterns induced by temperature anomalies and buoyancy fields. Furthermore, our simulations of large-scale localized temperature anomalies reveal insights into the impact of perturbation strength, size, and vertical structure on the evolution of eddy heat fluxes, including poleward heat flux, energy, and meridional elongation of the buoyancy field. We observe the initiation of atmospheric instability, leading to precipitation systems, such as rain bands, and asymmetric latent heat release due to moist convection in diabatic environments. This study identifies distinct patterns, including the formation of a comma cloud pattern in the upper troposphere and a comma-shaped buoyancy anomaly in the lower layer, accompanied by the emission of inertia gravity waves. Additionally, the role of buoyancy anomalies in generating heatwaves and precipitation patterns is emphasized, particularly in mid-latitude regions.

In summary, Aeolus 2.0, with its specific capabilities, contributes to our understanding of the complex interactions of moist convection, buoyancy anomalies, and atmospheric dynamics, shedding light on the dynamics of extreme weather events and their implications for climate studies.

References

1. Rostami, M., Zhao, B., & Petri, S. (2022). On the genesis and dynamics of MaddenJulian oscillation-like structure formed by equatorial adjustment of localized heating. Quarterly Journal of the Royal Meteorological Society, 148 (749), 3788-3813. Retrieved from https://rmets.onlinelibrary.wiley.com/doi/abs/10.1002/qj.4388 doi: https://doi.org/10.1002/qj.4388

2. Rostami, M., Severino, L., Petri, S., & Hariri, S. (2023). Dynamics of localized extreme heatwaves in the mid-latitude atmosphere: A conceptual examination. Atmospheric Science Letters, e1188. Retrieved from https://rmets.onlinelibrary.wiley.com/doi/abs/10.1002/asl.1188 doi: https://doi.org/10.1002/asl.1188

How to cite: Rostami, M. and Petri, S.: Exploring Extreme Weather and Climate Events with Aeolus 2.0: A Multi-layer moist-convective Thermal Rotating Shallow Water (mcTRSW) Dynamical Core, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2867, https://doi.org/10.5194/egusphere-egu24-2867, 2024.

Extreme value theory provides a universal limit for the extremes of continuous independent and identically distributed random variables and has proven to be robust to generalisation to wider classes of random variables, including stationary processes, some nonstationary processes and even trajectories on deterministic chaotic systems. This universality, together with the fact that these methods require data from only one realization of the system, has been exploited in applications to study many series of climate data.

Fitting a probability distribution to the extreme events of a data series generated by a chaotic dynamical system gives us not only probabilistic predictions of the intensity and return time of the events themselves, but also geometrical information about the local structure of the attractor and the predictability and persistence of the extreme events.

However, these methods are sensitive to the mathematical properties of the dynamical system that generates the data, and are seldomly even mentioned when they are applied to real climate data. One further caveat of these methods is that they are hard to falsify, i.e. we cannot verify easily if an answer is wrong. For these reasons, we explore how these methods respond to different systems with different complexity and different mathematical properties, trying to understand which of the results on the literature could meaningful and which could be numerical artifacts.

How to cite: del Amo, I., Datseris, G., and Holland, M.: Extreme value methods in dynamical systems of different complexity, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4539, https://doi.org/10.5194/egusphere-egu24-4539, 2024.

Within the climate model hierarchy, simple models usually play the important role of highlighting dynamical processes that can possibly govern climate phenomena. If, in addition, their results are in significant agreement with observations, the processes thus identified are even more likely to regulate the actual phenomena. In this context, the dynamical process of intrinsic variability paced by a deterministic forcing (also called deterministic excitation, DE [1]) is highlighted here by two simple models of different degrees of complexity and set in the different contexts of paleoclimate and physical oceanography. In both cases, despite the simplicity of the models, the results show significant agreement with observations.

The DE mechanism requires the system (i) to possess intrinsic nonlinear relaxation oscillations (ROs) and (ii) to be in the excitable state (i.e., ROs do not emerge spontaneously but can be excited, and therefore paced, by a suitable forcing); moreover, (iii) ROs are excited by a deterministic forcing if a given tipping point is passed.

In the first case [1], the abrupt late Pleistocene glacial terminations are shown by a conceptual model to correspond to the excitation, by the astronomical forcing, of ROs describing glacial-interglacial transitions (e.g., [2]). In the second case [3], ROs describing the Kuroshio Extension low-frequency variability [4] are shown, by a primitive equation ocean model, to be excited remotely by the North Pacific Oscillation. These results show how simple modeling approaches of different complexity advance process understanding and can, therefore, provide theoretical guidelines for interpreting state-of-the-art ESM results.

[1] Pierini S., 2023: The deterministic excitation paradigm and the late Pleistocene glacial terminations. Chaos, 33, 033108.

[2] Gildor H. and E. Tziperman, 2001: A sea ice climate switch mechanism for the 100-kyr glacial cycles. J. Geophys. Res., 106, 9117–9133.

[3] Pierini S., 2014: Kuroshio Extension bimodality and the North Pacific Oscillation: a case of intrinsic variability paced by external forcing. J. Climate, 27, 448-454.

[4] Pierini S., 2006: A Kuroshio Extension System model study: decadal chaotic self-sustained oscillations. J. Phys. Oceanogr., 36, 1605-1625.

How to cite: Pierini, S.: Simple oceanographic and paleoclimate modeling highlights the same dynamical process: intrinsic variability paced by a deterministic forcing, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4679, https://doi.org/10.5194/egusphere-egu24-4679, 2024.

The Amazon rainforest, one of the most important biomes in the world, and recognized as a potential tipping element in the Earth system, has received increasing attention in recent years. Theory and observations suggest that regional climate change from greenhouse gas emissions and deforestation may push the remaining forest toward a catastrophic tipping point.

Despite the urgency to assess the future fate of the Amazon, it remains unclear if state-of-the-art Dynamic Global Vegetation Models (DGVMs) can capture the highly nonlinear dynamics underlying such potentially abrupt dynamics and there is a noticeable scarcity of DGVM evaluations regarding their potential to predict forthcoming tipping points.

In our manuscript, we systematically investigate how the Amazon forest responds in idealized scenarios where precipitation is linearly decreased and subsequently increased between current levels and zero, using the state-of-the-art model LPJmL. We investigate whether large-scale abrupt changes and tipping points occur, and whether early warning signals as expected from theory can be detected. 

Our results indicate a pronounced nonlinearity but reversible behavior between vegetation aboveground biomass (AGB) and mean annual precipitation (MAP) in the LPJmL simulations. In particular, there exists a threshold at a critical rainfall level below which there is a rapid decrease in forest biomass. The value of the threshold is determined by seasonality, evapotranspiration and the adaptive capacity of roots. Significant "early warning signs" can be detected before the transition.

How to cite: Nian, D., Bathiany, S., Sakschewski, B., Drüke, M., Blaschke, L., Ben-Yami, M., von Bloh, W., and Boers, N.: The critical precipitation threshold for the Amazon forest biomass in the LPJmL vegetation model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5570, https://doi.org/10.5194/egusphere-egu24-5570, 2024.

The subpolar gyre (SPG) is one of the climate tipping elements which could have a large impact on the climate in the northern hemisphere. Improving our understanding of its dynamics is key to assessing the likelihood of it passing a tipping point. Some CMIP6 models exhibit abrupt transitions in the sea surface temperature in the SPG region, but the majority does not. The differences in the model response can be related to the stratification bias, with many models having a too strong stratification preventing them from exhibiting bistable gyre dynamics.

To better understand the SPG we study the (lagged) partial correlations between the relevant aspects of its dynamics in the CMIP6 ensemble. In contrast to standard correlations, partial correlations correct for the effect of autocorrelation and the effect of (the past of) other relevant variables. Therefore, it gives a better indication of there being a causal relation. Based on the partial correlation between the sea surface temperature and mixed layer depth we split the ensemble into two groups (strong or negligible relation) and for each select one model to study its dynamics in detail. In addition, we discuss the interaction of the SPG with the Atlantic Meridional Overturning Circulation (AMOC) using the same methods. These results can help in better informing more conceptual climate models of the SPG, AMOC and their interactions, which can be used to study potential tipping dynamics.

How to cite: Falkena, S. and von der Heydt, A.: Disentangling the dynamics of the subpolar gyre and its interaction with the AMOC in the CMIP6 ensemble, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8104, https://doi.org/10.5194/egusphere-egu24-8104, 2024.

Complex numerical models of climate consist of simulation of fluid dynamics and thermodynamics on a discrete grid, and parameterizations, which are algorithms that approximate processes smaller than gridscale. Because parameterizations of a given process may be written as different functions of different, potentially non-observable variables, it can be difficult to quantify the process differences between individual climate models and between climate models and the real world.

Here, we attempt to write down a simple linear model that represents the response of the Earth's tropical land surface to atmospheric forcing on monthly timescales in terms of the same observable variables using a technique called continuous structural parameterization. Simulated data are taken from complex General Circulation Models (GCMs) run under the AMIP protocol and a CESM2 perturbed physics ensemble (PPE) of our own devising;  observed measurements are taken from FLUXNET flux tower sites. We find that the simple model captures land surface behaviour well except in mountainous regions.

Establishing a generalised parameter space, we see that most GCMs are in reasonable agreement with FLUXNET at FLUXNET sites, although there is evidence that GCMs consistently slightly overestimate the response of surface turbulent fluxes to downward radiation. Further, it is found that the differences between structurally different AMIP models are considerably greater than the differences between CESM2 PPE members -- even though the PPE parameters are varied across their realistic domain. If the simple model is trained only at GCM spatial gridpoints that contain a FLUXNET site, there is little degradation in simple model performance compared with global training, suggesting that even the few available tropical FLUXNET sites are useful for constraining land surface model response throughout the tropics. This is of course contingent on whether or not point measurements taken by FLUXNET are representative of the wider area around FLUXNET sites.

How to cite: Lambert, F. H., Zarakas, C., Natchiar S. R., M., Swann, A. L. S., and Koven, C. D.: Using a simple model to measure the differences between climate model land surface simulations and FLUXNET observations, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8117, https://doi.org/10.5194/egusphere-egu24-8117, 2024.

The Atlantic Meridional Overturning Circulation (AMOC) and polar ice sheets are considered susceptible to critical transitions under climate change. Identified as core tipping elements, their collapse would have global and drastic consequences. Furthermore, the AMOC and polar ice sheets form a complex interacting system, where the collapse of one component can heavily impact the stability of others. In the worst case, this could result in a large-scale domino effect, otherwise known as a cascading tipping event.

In this presentation, our focus is on assessing the stability of the AMOC in the presence of tipping Greenland ice sheet (GIS) and West Antarctica ice sheet (WAIS). While most existing studies agree on the destabilizing impact of a GIS collapse on the AMOC, the consequences of a WAIS collapse remain uncertain. A previous conceptual study suggested that a WAIS tipping event might actually prevent an AMOC collapse against both climate warming and increased GIS meltwater fluxes. Using a better conceptual model of the AMOC, we demonstrate that both the melting rate and natural variability associated with surface meltwater fluxes are decisive factors for this phenomenon to occur. Finally, we present preliminary findings in which the relevance of this stabilizing effect is investigated in the model of intermediate complexity CLIMBER-X.

How to cite: Sinet, S., von der Heydt, A. S., Ashwin, P., and Dijkstra, H. A.: AMOC Stability amid Tipping Ice Sheets from Conceptual to Intermediate Complexity Models, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8397, https://doi.org/10.5194/egusphere-egu24-8397, 2024.

The Atlantic Meridional Overturning Circulation (AMOC) exerts a major influence on global climate. There is much debate about whether the current strong AMOC may collapse as a result of anthropogenic forcing and/or natural variability. Here, we ask whether internal decadal variability could affect the likelihood of AMOC collapse. We examine natural variability of basin-scale salinities and temperatures in four CMIP6 pre-industrial runs. We fit the CMIP6 variability to several empirical, linear noise models, and to a nonlinear, process-based AMOC model. The variability is weak and its processes inconsistent among the CMIP6 models considered. Based on the CMIP6 variability levels we find that noise-induced AMOC collapse is unlikely in the pre-industrial climate, but plausible if external forcing has shifted the AMOC closer to a threshold, which can be identified for the non-linear model using bifurcation analysis. However the CMIP6 models may systematically underestimate current Atlantic Ocean variability, and we find that substantially stronger variability would increase the likelihood of noise-induced collapse.

How to cite: Chapman, R., Ashwin, P., Wood, R., and Baker, J.: Quantifying risk of a noise-induced AMOC collapse from northern and tropical Atlantic Ocean variability, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8729, https://doi.org/10.5194/egusphere-egu24-8729, 2024.

Reduced order quasi-geostrophic land-atmosphere coupled models display qualitatively realistic mid-latitude atmosphere behaviour, meaning that such models can produce typical atmospheric dynamical features such as atmospheric blocking. At the same time, due to a low number of degrees of freedom, they are still simple enough to allow for analysis of the system dynamics. These features mean that these models are well suited to investigating bifurcations in atmospheric dynamics, and use a dynamical systems approach to better understand the corresponding atmospheric behaviour. 

This project introduces a symbolic python workflow for using the flexible  land-atmosphere (qgs, 2020) spectral model with the continuation software AUTO. This work builds on the results of Xavier et al. (2023) to understand how the model variability and predictability is impacted by the model resolution. We also use bifurcation diagrams to better understand how parameters such as atmosphere-land friction impact the atmospheric blocking, and in turn the model atmosphere predictability. This is done for a range of model resolutions to investigate how the number of degrees of freedom impacts both the realism of the model, but also the structures found in the dynamics.

Demaeyer, Jonathan & De Cruz, Lesley & Vannitsem, S.: qgs: A flexible Python framework of reduced-order multiscale climate models. Journal of Open Source Software. 5. 2597. 10.21105/joss.02597, 2020. 

Xavier, A. K., Demaeyer, J., and Vannitsem, S.: Variability and Predictability of a reduced-order land atmosphere coupled model, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2023-2257, 2023.

How to cite: Hamilton, O., Demaeyer, J., Xavier, A., and Vannitsem, S.: The impact of model resolution on variability in a coupled land atmosphere model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9253, https://doi.org/10.5194/egusphere-egu24-9253, 2024.

Every year, the area of the Arctic sea-ice decreases in the boreal spring and summer and reaches its yearly minimum in the early autumn. Due to global warming, Arctic summer sea ice will most probably disappear. As the sea ice cover decreases, its border is retreating northwards towards the central Arctic. This retreat is not uniform in space and the variability of the border’s movement further North could yield an early warning signal for summer sea ice loss. Here, we track the sea ice border from time series obtained from models of various complexity and observations to study the spatial variability of the border’s movement as Arctic summer sea ice approaches its disappearance.

How to cite: Hummel, C.: Spatial fluctuations of the Arctic sea ice border, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10070, https://doi.org/10.5194/egusphere-egu24-10070, 2024.

Climate variability is crucial to our understanding of future climate change and its impacts on societies and the natural world. However, the climate records of the observational era are too short to explore long-term variability. Conversely, an exploration of long transient simulations from state-of-the-art Earth System Models (ESMs) poses high computational demands. It is therefore pertinent to identify the level of complexity sufficient to simulate the variability of surface climate from annual to centennial and longer timescales.

To this end, we use an ensemble of transient simulations of the Last Deglaciation, the last period of significant global warming. The ensemble covers an energy balance model (EBM), models of intermediate complexity (EMICs), general circulation models (GCMs) and ESMs. This constitutes a hierarchy that we categorize based on employed atmosphere and ocean components and their resolution, as well as implemented radiation, land hydrology, vegetation and aerosol schemes.

To investigate the simulated variability of surface temperature and precipitation, we analyze changes in the shapes of their distributions as characterized by their higher order moments – variance, skewness, kurtosis – with warming. These higher order moments relate the tails to the extremes of the distributions. We identify spatial and temporal patterns and how they depend on model complexity. The EMICs can generally match the global and latitudinal changes in temperature variability found in more complex models. However, they lack in precipitation variability. We further find that the EMICs fail to simulate the tails of the precipitation distributions. We observe dependency of variability on the background state, generally increasing with model complexity. However, there is still a large spread between models of similar complexity, some of which can be related to differences in forcings. Furthermore, questions remain on the abilities of models of any complexity to simulate a magnitude of long-term variability similar to that found regionally in proxy reconstructions. Our analysis offers implications as to the complexity needed and sufficient for capturing the full picture of climate change and we offer some first insights into how the findings translate to future projections of climate change.

How to cite: Ziegler, E., Weitzel, N., Baudouin, J.-P., Kapsch, M.-L., Mikolajewicz, U., Gregoire, L., Ivanovic, R., Valdes, P., Wirths, C., and Rehfeld, K.: Dependence of simulated variability of surface climate on model complexity – insights from an ensemble of transient simulations of the Last Deglaciation, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11984, https://doi.org/10.5194/egusphere-egu24-11984, 2024.

We report on the development of a novel process model created to study ocean-sea ice interaction and the dynamics of the upper ocean in the marginal ice zone (MIZ), built using the Community Earth System Model (CESM). Our model uses the MOM6 ocean model and CICE6 sea-ice model as active components within CESM, on a custom ~50km x ~50km grid with a horizontal resolution of ~50m, extending to a depth of 75m (30 vertical layers). The model allows for either reflecting or zonally re-entrant boundary configurations. Atmospheric forcing is imposed through a simplified data atmosphere component that provides constant forcing over the model domain. Results from several simple scenarios are presented and compared to results obtained using the MITgcm.

By working within CESM, we are able to leverage CESM’s existing infrastructure and capabilities, including the use of the Community Mediator for Earth Prediction Systems (CMEPS) for coupling between active components. Furthermore, additional model components that are already available within CESM (e.g., waves, atmosphere) can be incorporated into the process model in a straightforward way. Future work will include incorporation of a modified sea-ice component that allows tracking of individual floes utilizing a discrete element method approach.

How to cite: Hall, P., Horvat, C., Fox-Kemper, B., Brenner, S., and Altuntas, A.: A Novel Process Model of Ocean-Sea-Ice Interaction Using CESM, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12421, https://doi.org/10.5194/egusphere-egu24-12421, 2024.

The climate system can be numerically represented by a set of physically-based dynamical equations whose solution requires substantial computational resources. This makes computationally efficient, low dimensional emulators that simulate trajectories of the underlying dynamical system an attractive alternative for model evaluation and diagnosis. We suggest that since such an emulator must adequately capture anomaly evolution, its construction should employ a grid search technique where maximum forecast skill determines the best reference model. In this study, we demonstrate this approach by testing different bases used to construct a Linear Inverse Model (LIM), a stochastically-forced multivariate linear model that has often been used to represent the evolution of coarse-grained climate anomalies in both models and observations. LIM state vectors are typically represented in a basis of the leading Empirical Orthogonal Functions (EOFs), but while dominant large-scale climate variations often are captured by a subset of these statistical patterns, key precursor dynamics involving relatively small scales are not. An alternative approach is balanced truncation, where the dynamical system is transformed into its Hankel space, whose modes span both precursors and their subsequent responses. Constructing EOF- and Hankel-based LIMs from monthly observed anomalous Pacific sea surface temperatures, both for the 150-yr observational record and a perfect model study using 600 yrs of LIM output, we find that no balanced truncation model of any dimension can outperform an EOF-based LIM whose dimension is chosen to maximize independent skill. However, the dynamics of a high-dimensional EOF-based LIM can be efficiently reproduced by far fewer Hankel modes.

How to cite: Xu, T., Newman, M., Alexander, M., and Capotondi, A.: A Forecast Test for Reducing Dynamical Dimensionality of Model Emulators, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-14067, https://doi.org/10.5194/egusphere-egu24-14067, 2024.

We introduce a stochastic, energy-balance, climate model defined over the macroweather regime (approximately 15 days or longer). Together, the energy balance principle, combined with the model’s natural scaling, demonstrate quite promising results despite the relative simplicity. A special case of the model can also be derived from a very classical basis, and, because of some similarities, we propose this model as a development upon the work of Budyko and Sellers.

When the classical Budyko-Sellers energy balance model is updated by using the (correct) radiative-conductive surface boundary conditions, one obtains the Fractional Energy Balance Equation (FEBE). The FEBE involves fractional space-time operators and its generic solutions are scaling, in agreement with much atmospheric and oceanic data. In time, it implies long range memories that have been successfully used to make both multi-decadal climate projections as well as monthly and seasonal (long range) forecasts. In space, the FEBE is nonlocal so that energy flux imbalances at any location can affect the balance in locations far away. This is possible because the model operates over monthly and longer time scales; over these scales, energy can be both stored and transported in the atmosphere, ocean, and subsurface.

Until now, the FEBE’s full nonlocal space-time interaction operator has been only approximated. Here, by introducing a numerical model, the nonlocal dynamics of the FEBE and corresponding Earth-system FEBE energy flows over the 2D Earth surface are fully detailed.

We propose the FEBE as an alternative to more conventional, deterministic, weather-regime-based climate models. Given the generality of the ideas pursued here - the use of fractional operators; the use of stochasticity and the macroweather regime - there seems a great potential for these to be used much more widely. Hopefully this research, and possibly related works, will encourage a greater diversity of pursuits and be inspiring to others in their own work.

How to cite: lebiadowski, D. and Lovejoy, S.: Nonlocal energy fluxes and fractional operators in updated, stochastic, Budyko-Sellers models, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-14439, https://doi.org/10.5194/egusphere-egu24-14439, 2024.

A fully-coupled general circulation model of intermediate complexity is documented. The study presents an overview of the model climatology and variability, with particular attention for the phenomenology of processes that are relevant for the predictability of the climate system on seasonal-to-decadal time-scales. It is shown that the model can realistically simulate the general circulation of the atmosphere and the ocean, as well as the major modes of climate variability on the examined time-scales: e.g. El Niño-Southern Oscillation, North Atlantic Oscillation, Tropical Atlantic Variability, Pacific Decadal Variability, Atlantic Multi-decadal Variability. We demonstrate the ability of the model in simulating non-stationarity of coupled ocean-atmosphere modes of variability. Potential applications of the model are discussed, with emphasis on the possibility to generate sets of low-cost large-ensemble retrospective forecasts. We argue that the presented model is suitable to be employed in traditional and innovative model experiments that can play a significant role in future developments of seasonal-to-decadal climate prediction.

How to cite: Ruggieri, P., Abid, M. A., Garcia-Serrano, J., Grancini, C., Kucharski, F., Pascale, S., and Volpi, D.: SPEEDY-NEMO: performance and applications of a fully-coupled intermediate-complexity climate model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-14831, https://doi.org/10.5194/egusphere-egu24-14831, 2024.

Hasselmann’s paradigm, introduced in 1976 and recently honoured with the Nobel Prize, can, like many key innovations in the sciences of climate and complexity, be understood on several different levels, both technical and conceptual. It can be seen as a mathematical technique to add stochastic variability into pioneering energy balance models (EBMs) of Budyko and Sellers. On a more conceptual level, it used the mathematics  of Brownian motion to provide an  abstract superstructure linking slow climate variability to fast weather fluctuations, in a context broader than EBMs, leading Hasselmann to posit the need for negative feedback in climate modelling.

Hasselmann's paradigm itself has much still to offer us [e.g. Calel et al, Naure Communications, 2020], but naturally, since the 1970s a number of newer developments have built on his pioneering ideas. One important one has been the development of a rigorous mathematical hierarchy that embeds Hasselmann-type models in the more comprehensive Mori-Zwanzig (MZ) framework  (e.g.  Lucarini and Chekroun, Nature Reviews Physics, 2023). Another has been the interest in long range memory in stochastic EBMs, notably Lovejoy et al’s Fractional Energy Balance Equation [FEBE, discussed in this week’s Short Course SC5.15 ]. These have a memory with slower decay and thus longer range than the exponential form seen in Hasselmann’s EBM. My presentation [based on Watkins et al, in review at Chaos] attempts to build a bridge between MZ-based extensions of  Hasselmann, and the fractional derivative-based FEBE model.  I will argue that the Mori-Kubo overdamped Generalised Langevin Equation, as widely used in statistical mechanics, suggests the form of a relatively simple stochastic EBM with memory for the global temperature anomaly, and will discuss how this relates to FEBE.

How to cite: Watkins, N. W., Calel, R., Chapman, S., Chechkin, A., Klages, R., and Stainforth, D.: The Challenge of Non-Markovian Energy Balance Models in Climate, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-16958, https://doi.org/10.5194/egusphere-egu24-16958, 2024.

We introduce a minimal dynamical system derived from the classical Phillips two-level model with the goal of elucidating the essential mechanisms responsible for the interaction between eddies and mean flow. The choice of a two-level model as starting points allows for appreciating the relative role of barotropic and baroclinic processes. Specifically, we wish to explore the eddy saturation mechanism, whereby, when average conditions are considered, direct forcing of the zonal flow increases the eddy kinetic energy, while the energy associated with the zonal flow does not increase. The eddy-driven jet stream and storm tracks in the mid-latitude atmosphere are known to shift in latitude on various timescales, but the physical processes that cause these shifts are still unclear. Using our low-order model, we aim to understand the link between the structure of the eddies and the shift of the latitudinal maximum of the zonal flow in the mid-latitude atmosphere. Our findings elucidate the basic mechanisms behind baroclinic adjustment and provide insights into the properties of the storm track change between the jet entrance and jet exit regions of the North Atlantic.

How to cite: Lucarini, V., Kobras, M., and Ambaum, M.: Eddy Saturation and Latitudinal Storm Track Shift in a Reduced Two-level Model of the Atmosphere, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18360, https://doi.org/10.5194/egusphere-egu24-18360, 2024.

We utilize ideas from signal processing to demonstrate a novel methodology for climate emulation based on the response of the climate system to effective radiative forcing (ERF). While previous work has demonstrated the efficacy of impulse response functions as a tool for climate emulation, these methods are largely non-generalizable to new scenarios and are inaccessible to more general audiences. To remedy this, we propose a generalizable framework for emulation of climate variables such as near-surface air temperature, representing the climate system through the surrogate of spatially resolved impulse response functions. These response functions are derived through the deconvolution of ERF and near-surface air temperature profiles, treating ERF and near-surface air temperature as input and output signals, respectively. Using this framework, new scenarios can be quickly and easily emulated through convolution and other sets of impulse response functions can be derived from any pair of climate variables. We present results from an application to near-surface air temperature based on ERF and temperature data taken from experiments in the sixth phase of the Coupled Model Intercomparison Project (CMIP6). We evaluate the emulator using additional experiments taken from the CMIP6 archive, including the Shared Socioeconomic Pathways (SSPs), demonstrating accurate emulation of global mean and spatially resolved temperature change with respect to the outputs of the CMIP6 ensemble. Global absolute error in emulated temperature averages 0.25 degrees Celsius with a bias ranging from -0.14 to -0.04 degrees Celsius. We additionally show how our emulator can be implemented as a tool for climate education through integration with the En-ROADS platform, providing fast visualizations of spatially resolved temperature change for a number of policy-relevant scenarios. While it is unable to capture state-dependent climate feedbacks, such as the non-linear effects of Arctic sea ice melt in high-warming scenarios, our results show that the emulator is generalizable to any scenario independent of the specific forcings present.

How to cite: Womack, C., Eckley Selin, N., and Eastham, S.: Rapid Emulation of Spatially Resolved Temperature Response Functions to Effective Radiative Forcing, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18412, https://doi.org/10.5194/egusphere-egu24-18412, 2024.

The evolution of the Earth’s climate from the past to the future is explored by a hierarchy of models ranging from conceptual models to full-complexity, high-resolution Earth System Models (ESMs) (Claussen et al., 2002). The strength of conceptual models lies in the clarity of representing the concept of interactions between different climate processes, while ESMs  offer greater realism when it comes to spatial or temporal detail. Intermediate complexity models are somewhere in between, they are able to provide a big picture for long timescales. A common pattern throughout the model hierarchy, except for conceptual models illustrating multiple steady states, is often linearity of model responses to external forcing. This linearity can be visible in transient experiments, but also in equilibrium simulations. The question arises: is this linearity an artefact of our models, or is it reflective of reality?

In most cases, the linear response is likely representative of reality. As an example, we will focus on the linearity of land-related processes, such as climate-carbon feedbacks and permafrost-hydrology interactions. Permafrost systems have thresholds at 0°C, leading to nonlinearities at the local scale, but the combined response at large spatial scales tends to be more linear. However, nonlinear and abrupt changes are evident in geological records. For instance, the abrupt onset of the Bölling/Alleröd warming about 14.8 thousand years ago indicates that nonlinear changes on large spatial scales are indeed a real, albeit very rare, phenomenon. We will discuss possible reasons for the predominant linearity of the models and explore whether high-resolution models might show more nonlinear responses than coarse-grid models.

Reference:

Claussen, M., Mysak, L., Weaver, A. et al. Earth system models of intermediate complexity: closing the gap in the spectrum of climate system models. Climate Dynamics 18, 579–586 (2002). https://doi.org/10.1007/s00382-001-0200-1

How to cite: Brovkin, V., Stacke, T., de Vrese, P., Kleinen, T., and Winkler, A.: From conceptual to complex earth system models: why are models so linear?, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-22102, https://doi.org/10.5194/egusphere-egu24-22102, 2024.

Sea-ice ridges form as a part of sea-ice deformation, while the ice is moved by winds and ocean currents. While ridging is a localized process, it is assumed to limit the compressive strength of sea ice in large scale. However, formulations of large-scale ice strength, as used in Earth System Models, do not consider individual ridge formation processes in detail. Thus, it is necessary to understand the energy spend in ridge formation and various processes related to generating ice rubble and redistributing it. To investigate ridge formation in detail, we use the Aalto University in-house discrete-element-method (DEM) model. This three-dimensional DEM model features deformable, multi-fracturing, ice floes, which can fail and form ridges when coming into contact, while recording the ridging forces. With this, we discuss why three-dimensional simulations are important to investigate ridge formation process.

How to cite: Muchow, M. and Polojärvi, A.: Ridge-formation simulations in three dimensions using discrete element methods, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-348, https://doi.org/10.5194/egusphere-egu24-348, 2024.

How to cite: Huan, X., Wang, J., and Liu, Z.: Monthly Arctic sea ice prediction based on a data-driven deep learning model , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1671, https://doi.org/10.5194/egusphere-egu24-1671, 2024.

In an idealised setup, a dynamics-only sea ice model is used to investigate the fully multivariate state and parameter estimations that uses a novel Maxwell-Elasto-Brittle (MEB) sea ice rheology. In the fully multivariate state estimation, the level of damage, internal stress and cohesion are estimated along with the observed sea ice concentration, thickness and velocity. In the case of parameter estimation, we estimate the air drag coefficient and the damage parameter of the MEB model. The air drag coefficients adjust the strength of the forcing on the sea ice dynamics while the damage parameter controls the mechanical behaviour of the internal property of sea ice. We show that, with the current observation network, it is possible to improve all model state forecast and the parameter accuracy using data assimilation approaches even though problems could arise in such an idealised setup where the external forcing dominates the model forecast error growth.

How to cite: Chen, Y., Smith, P., Carrassi, A., Pasmans, I., Bertino, L., Bocquet, M., Finn, T. S., Rampal, P., and Dansereau, V.: Multivariate state and parameter estimation using data assimilation in a Maxwell-Elasto-Brittle sea ice model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2377, https://doi.org/10.5194/egusphere-egu24-2377, 2024.

The coastal polynyas of the Southern Ocean play a crucial role in the formation of dense water and have an impact on the stability of ice shelves. Therefore, it is important to accurately simulate them in climate models. To achieve this goal, the relationship between grounded icebergs, landfast ice and polynyas appears to be central. Indeed, grounded icebergs and landfast ice are believed to be key drivers of coastal polynyas. However, ESMs do not simulate Antarctic landfast ice. Moreover, at a circumpolar scale, there are no observations of grounded icebergs available. Hence, we must seek model representations that can overcome these issues. To address these gaps, we conducted a study using an antarctic circumpolar configuration of the ocean–sea ice model NEMO4.2-SI–3 at the 1/4° resolution. We ran two simulations for the period 2001–17, with the only difference being the inclusion or exclusion of landfast ice information based on observations. All other factors, including initial conditions, resolution and atmospheric forcings, were kept the same. We then compared the results of these simulations with observations from the advanced microwave scanning radiometer to evaluate the performance of the new simulation. Our analysis allowed us to determine the extent to which prescribing the distribution of landfast ice and setting the sea ice velocity to zero on landfast ice regions influenced various aspects of the sea ice, such as polynyas, landfast ice and sea ice distribution in the model. In the future, we plan to look at the impact on the ocean and to develop a physical parameterization in order to model landfast ice and consequently polynyas on a permanent basis.

How to cite: Pirlet, N., Fichefet, T., Vancoppenolle, M., Rousset, C., Mathiot, P., Fraser, A., Barthélemy, A., and Kittel, C.: Perscribing Antarctic landfast sea ice in a sea ice-ocean model., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3367, https://doi.org/10.5194/egusphere-egu24-3367, 2024.

Ice mélange (a mixture of sea ice, bergy bits and icebergs) can have a strong influence on the sea-ice-ocean interaction. So far, ice mélange is not represented in climate models as numerically efficient realizations are missing. This motivates the development of an ice-mélange model based on the viscous-plastic sea-ice rheology, which is currently the most commonly used material law for sea ice in climate models. Starting from the continuum mechanical formulation, we modify the rheology so that icebergs are represented by thick, highly compact pieces of sea ice. These compact pieces of sea ice are held together by a modified tensile strength in the material law. In this framework, the ice mélange is considered as one single fluid, where the icebergs are realised by particles.
Using idealized test cases, we demonstrate that the proposed changes in the material law are crucial to represent icebergs with the viscous-plastic rheology. Similar to the viscous-plastic sea-ice model, the ice-mélange model is highly nonlinear. Solving the model at the resolution needed to represent the typical size of icebergs in ice mélange (< 300m) is therefore challenging. We show that the ice-mélange formulation can be approximated efficiently with a modified Newton's method. Overall, the simple extension of the viscous-plastic sea-ice model is a promising path towards the integration of ice mélange into climate models.

How to cite: Kahl, S. and Mehlmann, C.: A model for ice-mélange based on particle and continuums mechanics, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4144, https://doi.org/10.5194/egusphere-egu24-4144, 2024.

Rapid decline of Arctic sea ice has created more open water for ocean wave development and highlighted the importance of wave-ice interactions in the Arctic. Some studies have made contributions to our understanding of the potential role of the prognostic floe size distribution (FSD) on sea ice changes. However, these efforts do not capture the full interactions between atmosphere, ocean, wave, and sea-ice. In this study, a modified joint floe size and thickness distribution (FSTD) is implemented in a newly-developed regional atmosphere-ocean-wave-sea ice coupled model and a series of pan-Arctic simulation is conducted with different physical configurations related to FSD changes, including FSD-fixed, FSD-varied, lateral melting rate, wave-fracturing formulation, and wave attenuation rate. Firstly, atmosphere-ocean-wave-sea ice coupled simulations show that the prognostic FSD leads to reduced ice area due to enhanced ice-ocean heat fluxes, but the feedbacks from the atmosphere and the ocean partially offset the reduced ice area induced by the prognostic FSD. Secondly, lateral melting rate formulations do not change the simulated FSD significantly, but they influence the flux exchanges across atmosphere, ocean, and sea-ice and thus sea ice responses. Thirdly, the changes of FSD are sensitive to the simulated wave parameters associated with different wave-fracturing formulations and wave attenuation rates, and the limited oceanic energy imposes a strong constraint on the response of sea ice to FSD changes. Finally, the results also show that wave-related physical processes can have impacts on sea ice changes with the constant FSD, indicating the indirect influences of ocean waves on sea-ice through the atmosphere and the ocean.

How to cite: Yang, C.-Y. and Liu, J.: Understanding influence of ocean waves on Arctic sea ice simulation: A modeling study with an atmosphere-ocean-wave-sea ice coupled model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4502, https://doi.org/10.5194/egusphere-egu24-4502, 2024.

How to cite: Brivoal, T., Guemas, V., Rousset, C., and Vancoppenolle, M.: Improving the representation of snow over sea-ice in the SI3 model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5177, https://doi.org/10.5194/egusphere-egu24-5177, 2024.

Marginal ice zones are regions where individual sea ice floes interact mechanically and thermodynamically with turbulent ocean currents at the (sub-)mesoscale. Fine scale exchanges of momentum, heat and salinity at the interface between the ocean and the sea ice floes have important effects on upper-ocean energetics, under-ice tracer mixing, and the ice-pack melt rates. The dynamics of these moving floes remain poorly constrained, notably due to the challenge of numerically resolving sub-mesoscale processes and modelling the discrete behavior of sea ice in traditional climate models. 

Here, we use oceanic Large Eddy Simulations (LES), two-way coupled to a Discrete Element Model (DEM) of disk-shaped sea ice floes, to quantify the kinetic energy transfers between ocean and sea ice during summer-like conditions, varying sea ice concentration and floe size distribution. The damping of oceanic currents by floes is found to be important for a sea ice concentration as low as 40%, when the sizes of floes are comparable to the characteristic eddy size. This damping is largely compensated by the generation of kinetic energy due to melt-induced baroclinic instability at the edge of sea ice floes, leading to a net energy sink of approximately 15%, relative to a simulation with no floes. At higher sea ice concentrations, the oceanic kinetic energy production weakens, while energy loss due to ice/ocean damping and floe-floe collisions both increase. These energy fluxes are mediated by the spatial aggregation of sea ice floes that occurs within the high-strain regions surrounding ocean mesoscale eddies. Eddy-driven aggregation can also reduce the melt rate of small floes as they become shielded from warm waters by neighboring larger floes. These results highlight the need for scale-aware, and specifically floe-scale parameterizations of sea ice and its coupling to ocean turbulence, within global climate models.

How to cite: Gupta, M., Thompson, A., and Klein, P.: Floe-scale ocean / sea ice energy transfers in the marginal ice zone, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5374, https://doi.org/10.5194/egusphere-egu24-5374, 2024.

The behavior of sea ice has been studied for many decades. In order to model its viscous-plastic behavior at scales spanning several thousand kilometers, different numerical models have been proposed. Based on the established approach in [1], this contribution presents a simulation model for sea ice dynamics to describe the sea ice circulation and its evolution over one seasonal cycle. In course of that, the sea ice concentration and the sea ice thickness are considered, of which the physical behavior is governed by transient advection equations. Here, the sea ice velocity serves as coupling field.

Recently developed approaches base on a finite element implementation choosing a (mixed) Galerkin variational approach, see e.g. [2] and [3]. But therein, challenges may occur regarding the stability of the numerically complex scheme, especially when dealing with the first-order advection equations. Thus, we propose the application of the mixed least-squares finite element method, which has the advantage to be also applicable to first-order systems, i.e., it provides stable and robust formulations even for non-self-adjoint operators, such as the tracer equations (for sea ice thickness and sea ice concentration).

For solving the instationary sea ice equation the presented least-squares finite element formulation takes into account the balance of momentum and a constitutive law for the viscous-plastic flow. The considered primary fields are the stresses σ, the velocity v, the concentration A_ice and the thickness H_ice. In relation, four residuals are defined for the derivation of a first-order least-squares formulation based on the balance of momentum, the constitutive relation for the stresses, and two tracer-equations. Different approaches can be made with respect to the approximation functions of the primary fields, i.e., choosing e.g. conforming (H(div) interpolation functions) or non-conforming (Lagrangian interpolation functions) stress approximations, while Lagrangian interpolation functions are chosen for the remaining fields. In order to compare such approaches, the box test case is utilized, cf. [3], which is well described in literature.

References:

[1] W.D. Hibler III. A dynamic thermodynamic sea ice model. Journal of Physical Oceanography, 9(4):815-846, 1979.

[2] S. Danilov, Q. Wang, R. Timmermann, M. Iakovlev, D. Sidorenko, M. Kimmritz, T. Jung. Finite-Element Sea Ice Model (FESIM), Version 2. Geoscientific Model Development, 8:1747-1761, 2015.

[3] C. Mehlmann and T. Richter. A modified global Newton solver for viscous-plastic sea ice models. Ocean Modelling, 116:96-107, 2017.

How to cite: Hellebrand, S., Schwarz, C., and Schröder, J.: Application of mixed least-squares FEM to study sea ice dynamics, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6441, https://doi.org/10.5194/egusphere-egu24-6441, 2024.

Sea ice floe size can impact several processes that determine the evolution of the Arctic sea ice, including lateral melt volume, momentum exchange, and rheology. Floe size distribution (FSD) models are applied within continuum sea ice models to capture the evolution of the FSD through parameterisations of the processes that modify floe size such as lateral melting and wave break-up of floes. FSD models do not yet adequately resolve in-plane fragmentation processes of floes such as the breakup of floes under wind forcing, through interactions between neighbouring floes, or through thermal weakening. It is challenging to characterise and therefore parameterise these in-plane floe breakup processes due to limited availability of in-situ observations. Discrete element models (DEMs) offer an alternative way to understand the different mechanisms of floe fragmentation. By resolving relevant properties such as shear and normal stress and sea ice strength at the sub-floe scale, it is possible to use DEMs as a virtual laboratory and directly simulate the break-up of floes into smaller fragments.

In this study, we describe how in-situ observations of sea ice can be combined with output from sea ice DEMs to develop parameterisations of in-plane breakup of floes that can then be applied in continuum models. We then discuss the necessary model developments in order to apply a sea ice DEM to floe fragmentation at smaller scales. We will also present results from a series of DEM simulations used to model the fracture of sea ice under different forcing conditions and with varying sea ice states to identify the important sea ice parameters and processes in determining the size of the floes that form from in-plane breakup events.

How to cite: Bateson, A., Feltham, D., Schröder, D., Durski, S., Hutchings, J., Basu, R., and Hwang, B.: Using discrete element methods to understand in-plane fragmentation of sea ice floes, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6569, https://doi.org/10.5194/egusphere-egu24-6569, 2024.

Arctic sea ice has experienced a differential decline in speed due to the same anthropogenic greenhouse gas forcing, as evidenced by rapid decline after the end of the last century. Our convergent observations, last-millennium reanalysis, and model analyses have revealed that large tropical volcanic eruptions can lead to a decadal increase in Arctic sea ice, and the 1982 and 1991 large volcanic eruptions slowed down the decline of Arctic sea ice during the last century. The models, selected based on the observed sensitivity of Arctic sea ice to volcanic eruptions, suggest that the earliest ice-free summer year in the Arctic will be around 2040 in high-emission sceneria of SSP585. These findings emphasized the crucial need to incorporate volcanic influences when projecting future Arctic changes amid global warming.

How to cite: Wang, X.: Historical volcanic eruptions slowed down rapid decline in Arctic sea ice linked to global warming, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9554, https://doi.org/10.5194/egusphere-egu24-9554, 2024.

The rapidly evolving sea ice cover requires novel modeling approaches and mathematical techniques to accurately simulate the sea ice dynamics, thermodynamics, and its interactions with the atmosphere and ocean at varying spatiotemporal resolutions. In this context, the CMCC is developing the Multiscale Unstructured model for Simulating the Earth’s water environment (MUSE), a novel global ocean-sea ice model on unstructured meshes.

MUSE employs a finite-element numerical discretization on unstructured meshes, aiming at offering flexibility in simulating the global ocean for various applications, ranging from physical process understanding to operational sea ice predictions. The ongoing implementation of the sea ice component utilizes the traditional continuous sea ice formulation and the 2+1 split assumption, meaning that the sea ice dynamics and advection are solved for horizontal motions while the thermodynamics and radiative processes are parameterized at the subgrid scale.   

MUSE employs a modified elastic-viscous-plastic (mEVP) solver for the sea ice dynamics and a Flux Corrected Transport (FCT) advection scheme, alongside the state-of-the-art column physics package "Icepack" maintained by the CICE consortium.

Here, we describe the global implementation of the sea ice component in MUSE and its coupling with the ocean. We present the resulting representation of vertical thermodynamic processes and horizontal dynamics of sea ice.

How to cite: Cocetta, F., Zampieri, L., and Iovino, D.: The sea ice component of MUSE, the unstructured-mesh global ocean model of CMCC, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9802, https://doi.org/10.5194/egusphere-egu24-9802, 2024.

We showcase the Sea Ice Drift Forecast Experiment (SIDFEx) database. SIDFEx is a collection of close to 225,000 lagrangian drift forecasts for the trajectories of assets (mostly buoys) on the Arctic and Antarctic sea ice, at lead times from daily to seasonal with mostly daily resolution. The forecasts are based on systems with varying degrees of complexity, ranging from free-drift forecasts to forecasts by fully coupled dynamical general circulation models. Combining several independent forecasts allows us to construct a best-guess consensus forecast, with a seamless transition from systems with lead times of up to 10 days to systems with seasonal lead times. The forecasts are generated by 13 research groups using 23 distinct forecasting systems and sent regularly to the Alfred-Wegener-Institute, where they are archived and evaluated. Many groups send forecasts operationally in near-real time.

In our presentation, we will introduce the motivation behind and setup of SIDFEx, as well as an overview on the general forecast skill. We will focus on selected highlights, comprising the operational support of research cruises, short-term predictions of sea-ice deformation and regular contributions to the Sea Ice Outlook competition.

How to cite: Reifenberg, S. F., Ludwig, V., and Goessling, H. F. and the SIDFEx Team: Best of SIDFEx: Highlights and lessons learned from six years of sea-ice drift forecasting, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10098, https://doi.org/10.5194/egusphere-egu24-10098, 2024.

In this presentation we highlight recent developments in the implementation of Machine Learning (ML) algorithms into the large-scale sea ice model, SIS2. Specifically, we show how a Convolutional Neural Network (CNN) can be used to systematically reduce global sea ice biases during a 5-year ice-ocean simulation. The CNN has been trained to learn a functional mapping from model state variables to sea ice concentration Data Assimilation (DA) increments. Therefore, during model integration, the CNN ingests information about the numerical model's atmosphere, ocean, and sea ice conditions, and predicts the appropriate correction to the sub-grid category sea ice concentration terms (without seeing any actual sea ice observations). We also show how this combined DA+ML approach leads to a natural framework for augmenting training data for neural networks; one which can lead to significant improvements in online performance, without the need for direct online learning. The bias reductions over the 5-year simulation period for this CNN correction scheme are even competitive with the bias reductions achieved from DA. These findings therefore suggest that our approach could be used to reduce systematic sea ice biases in fully coupled climate model predictions on seasonal-to-climate timescales.

How to cite: Gregory, W., Bushuk, M., Zhang, Y., Adcroft, A., and Zanna, L.: Towards improving numerical sea ice predictions with data assimilation and machine learning, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11288, https://doi.org/10.5194/egusphere-egu24-11288, 2024.

The NEMO sea ice model, SI3, includes the simple formulation of Hibler (1979; H79) to parameterise the compressive strength of sea ice. This assumes that thick and compact sea ice has more strength than thin and low concentration sea ice. However, the H79 strength scheme does not consider physical assumptions around energy conservation. The strength scheme of Rothrock (1975; R75) is based on the amount of potential energy gained and frictional energy dissipated during ridging, and has been introduced to SI3. Additionally, the option for a negative exponential redistribution of ridged ice among thickness categories, to better approximate observations and improve stability compared to the existing uniform redistribution when using R75, has been included. The R75 strength formulation is stable and works well in SI3 at version 4.2 with an EVP rheology, under a Met Office forced NEMO/SI3 model configuration. Sea ice strength is generally reduced for the R75 scheme compared to H79. The most notable effect on the model output is a greater number of, and sharper, features in the resulting modelled ice field when using the R75 scheme compared to the H79 scheme, which are particularly apparent in the ice thickness field. An increase in the model effective resolution is therefore demonstrated.

How to cite: Fiedler, E., Blockley, E., Rousset, C., and Vancoppenolle, M.: Sea ice strength in SI3, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11413, https://doi.org/10.5194/egusphere-egu24-11413, 2024.

The current generation of sea-ice models with Brittle rheologies can represent the observed temporal and spatial scaling of the sea-ice dynamics at resolutions of around 10 km. However, running those models is expensive, which can prohibit their use in coupled Earth system models. The promising results of neural networks for the fast prediction of the sea-ice extent or sea-ice thickness offer an opportunity to remedy this shortcoming. Here, we present the development of a data-driven sea-ice model based on generative deep learning that predicts together the sea-ice velocities, concentration, thickness, and damage. Trained with more than twenty years of simulation data from neXtSIM, the model can extrapolate to previously unseen conditions, thereby exceeding the performance of baseline models.

Relying on deterministic data-driven models can lead to overly smoothed predictions, caused by a loss of small-scale information. This is why the ability to perform stochastic predictions can be instrumental to the success of data-driven sea-ice models. To generate stochastic predictions with neural networks, we employ denoising diffusion models. We show that they can predict the uncertainty that remains unexplained by deterministic models. Furthermore, diffusion models can recover the information at all scales. This resolves the issues with the smoothing effects and results in sharp predictions even for longer horizons. Therefore, we see a huge potential of generative deep learning for sea-ice modelling, which can pave the way towards the use of data-driven models within coupled Earth system models.

How to cite: Finn, T. S., Durand, C., Porro, F., Farchi, A., Bocquet, M., Chen, Y., and Carrassi, A.: A data-driven sea-ice model with generative deep learning, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11908, https://doi.org/10.5194/egusphere-egu24-11908, 2024.

There's increasing transpolar shipping in both the Arctic and Antarctic as a result of the reduction of sea ice and the desire from social economics.  Sea ice is a hazard for shipping in ice-infested water, Ship navigability in ice-covered sea depends on sea ice concentration, ice thickness, fraction of pressure ridges, and multi-year ice as well as ice speed and compression, it also depends on the vessel ice class. IMO introduced Risk Index Outcome(RIO) to provide guidelines for safe navigation, calculation of RIO requires accurate sea ice information including sea ice concentration and thickness. We developed a method similar to RIO to calculate navigation risk indicators using forecasting models including ECMWF S2S data, Copernicus data, and DMI data. Other than conventional sea ice parameters sea ice concentration and sea ice thickness, ice salinity, and ice age are also taken into account in risk indicator calculation. We select the time March 2019 -Oct 2020 and adopt the initial condition of the model forecast for sea ice to demonstrate the capabilities of seasonal forecasting of this navigation risk indicator in different models. In future, the calculation method will be implemented within the ClimateDT environment.

How to cite: Zhang, X.: Development of ship navigation risk indicator in sea ice-infested water, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12451, https://doi.org/10.5194/egusphere-egu24-12451, 2024.

The marginal ice zone is the transition region between the dense floating ice pack and the open ocean. In this zone, the interaction of surface waves with sea ice is highly complex. The sea ice is broken up into fragments, the floes, which can split into smaller parts and drift under the action of waves and underwater current. Although the downscaling is challenging, laboratory model experiments can contribute to a better understanding of this process coupling fluid and solid mechanics on a large range of time and space scales. We propose to study the fragmentation of a floating membrane, made up of 10 µm graphite particles arranged in a monolayer, by gravity surface waves with a wavelength of around 15 cm [1]. For a sufficiently strong wave amplitude, the raft progressively breaks up, developing cracks and producing fragments whose sizes decrease over a time scale that is long relative to the wave period. We then study the distribution of the fragments produced during the fragmentation process. The visual appearance of the size-distributed fragments surrounded by open water bears a striking resemblance to the floes produced by the fracturing of sea ice by waves. The fragmentation concepts and morphological tools developed for sea ice floes can be applied to our macroscopic analog. Although the mechanics of the two systems differ in their physical properties and in the fracture process, our experiment provides a model laboratory system for studying the fragmentation of floating 2D materials

[1] Saddier, L., Palotai, A., Aksil, M., Tsamados, M., & Berhanu, M. (2023). Breaking of a floating particle raft by water waves. In arXiv preprint arXiv:2310.16188.

How to cite: Berhanu, M., Saddier, L., Aksil, M., Ambre, P., and Tsamados, M.:  A laboratory model of fragmentation of a 2D membrane by waves. Analogies and differences with sea ice., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12804, https://doi.org/10.5194/egusphere-egu24-12804, 2024.

Because of a spring predictability barrier, the seasonal forecast skill of Arctic summer sea ice is limited by the availability of melt-season sea ice thickness (SIT) observations. The first year-round SIT observations, retrieved from CryoSat-2 from 2011 to 2020, are assimilated into the GFDL ocean–sea ice model. The model's SIT anomaly field is brought into significantly better agreement with the observations, particularly in the Central Arctic. Although the short observational period makes forecast assessment challenging, we find that the addition of May–August SIT assimilation improves September local sea ice concentration (SIC) and extent forecasts similarly to SIC-only assimilation. Although most regional forecasts are improved by SIT assimilation, the Chukchi Sea forecasts are degraded. This degradation is likely due to the introduction of negative correlations between September SIC and earlier SIT introduced by SIT assimilation, contrary to the increased correlations found in other regions.

How to cite: Zhang, Y.-F., Bushuk, M., Winton, M., Hurlin, B., Gregory, W., Landy, J., and Jia, L.: Improvements in September Arctic sea ice predictions via assimilation of summer CryoSat-2 sea ice thickness observations, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12912, https://doi.org/10.5194/egusphere-egu24-12912, 2024.

CMIP6 models present our best understanding of the Earth system, yet they currently fail to simulate a plausible evolution of sea ice area to changes in the global-mean temperature. We aim to assess whether correcting the temperature and Arctic Amplification biases between CMIP6 models and observations can simulate a sensitivity of sea ice loss to global warming that is within the plausible range. To do this, we develop an emulator that is calibrated to physically-based CMIP6 models and then constrained to observations. Such a tool efficiently translates the global-mean temperature of a specific year into a physically-based and observationally constrained probabilistic ensemble of SIA in each month. This setup allows our emulator to capture the core physical processes of CMIP6 projections, while capturing the observed sensitivity of sea ice loss to global warming through the observational constraint of Arctic Amplification. While there are many application possibilities of our emulator, we use our model here to probabilistically diagnose the timing of an ice-free Arctic Ocean. We find that under a high (SSP5-8.5), medium (SSP2-4.5) and low (SSP1-2.6) emission scenario, an ice-free September Ocean is ‘likely’ at 1.73 of global warming above the pre-industrial level, however we note that the probability in the lower emission scenario reduces to ‘unlikely’ in the late 21^st century as the global temperature partially recovers. Our projections suggest that the probability of an ice-free summer ocean rises rapidly from ‘unlikely’ at 1.5 of global warming to ‘likely’ at 2 of global warming, stressing the importance of preventing global temperatures rising above 1.5, as the probability of losing sea ice coverage in September rises sharply thereafter. For March, we also find that the observational constraints increase the probability of an ice-free ocean under SSP5-8.5, becoming ‘likely’ in early 2200, while the probability remains very low under SSP2-4.5 and SSP1-2.6 as less than 5% of models reach ice-free conditions. Our projections suggest an ice-free summer ocean could occur at 0.5 cooler levels than the CMIP6 multi-model ensemble mean implies. Likewise, our approach suggests the probability of an ice-free Arctic Ocean year-round is increased when constraining the Arctic Amplification to observations.

How to cite: Chilcott, S., Meinshausen, M., and Notz, D.: A MAGICC Arctic Sea Ice Emulator, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-13528, https://doi.org/10.5194/egusphere-egu24-13528, 2024.

Presently, climate models employ a continuum approach to describe sea ice. This approach assumes that statistical averages can be derived from a large number of ice floes. However, employing continuum rheological models at or below the scale of individual floes is only valid if the failure mode of a single floe aligns with that of an aggregate of floes. Initially, continuum models were designed for a grid resolution of 100 km. With recent advancements in computing power, sea-ice models are frequently operated at higher mesh resolutions, potentially leading to grid cells that no longer contains a representative sample of sea-ice floes.

We are addressing these shortcomings of current continuum sea-ice models by developing a hybrid model. The idea of the hybrid approach is to nest a particle model into a continuum sea-ice model in order to predict sea ice on fine spatial scales in a region of interest. An important component of particle models is a drag law to describe the influence of ocean and atmospheric currents on the floes. Measurements obtained onboard the Polarstern expedition PS 138 have shown that the correlation cannot be described fully locally, in regions with strongly heterogenous ice cover. Instead, larger surrounding flows have a substantial effect on the motion of small ones. Detailed numerical simulations of idealised test cases do confirm these findings.   

How to cite: Mehlmann, C. and Richter, T.: Calibration of a  hybrid sea-ice model based on particle and continuums mechanics, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15156, https://doi.org/10.5194/egusphere-egu24-15156, 2024.

Antarctic sea-ice variability affects the ocean and atmosphere both locally through thermodynamic processes and beyond the Antarctic regions remotely through dynamic processes, which may all change due to global warming. In this study, we develop the ANTSIC-UNet, a deep-learning model trained on physically enriched climate variables, to predict the extended seasonal Antarctic sea ice concentration of up to 6 months in advance. We assess the predictive skill of ANTSIC-UNet as regards linear trend prediction and anomaly persistence prediction in the Pan- and regional Antarctic areas using comparative analyses with two baseline models. Our results exhibit superior performance of ANTSIC-UNet for the extended seasonal Antarctic forecast. The predictive skill of ANTSIC-UNet is notably season-dependent, showing distinct variations across regions. Optimal prediction accuracy is found in winter, while diminished skill found during the summer can be largely attributed to the ice-edge error. High predictive skills are found in the Weddell Sea throughout the year, which suggests that regional Antarctic sea-ice predictions beyond 6 months are possible. We further quantify variable importance through a post-hoc interpretation method which indicates that ANTSIC-UNet has learned the relationships between SIC and other climate variables and the method therefore provides information on the physics of the model. At short lead times, on timescales of up to two months, ANTSIC-UNet predictions exhibit heightened sensitivity to sea surface temperature, radiation conditions and vertical atmospheric circulation conditions in addition to the sea-ice itself. At longer lead times, predictions are dependent on stratospheric circulation patterns at 7-8 months lead in addition to sea-ice. Furthermore, we discuss the potential of implementing physical constraints to enhance sea-ice-edge predictability.

How to cite: Yang, Z., Liu, J., and Grand Graversen, R.: Extended seasonal forecast of Antarctic Sea Ice using ANTSIC-UNet, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15487, https://doi.org/10.5194/egusphere-egu24-15487, 2024.