Sofia Vrettou; Theano Iliopoulou, and Demetris Koutsoyiannis

The tendency to avoid non-renewable sources in electricity production has led –among other alternative models of energy production- to the rapid growth of offshore wind farms. With regard to the structural design of the offshore wind turbines, international standards for the design of the turbines, suggest that the forces executed on the turbine’s pile, primarily induced by wind and waves, should be calculated by using dynamic models and short-term simulations of the natural parameters. However, due to the stochastic nature of wind and wave generated forces, such dynamic procedures and the short-term approach fail to accurately calculate or either - in the desired scenario - predict the exerted loads. On the other hand, stochastic methods do take into account characteristics of natural processes such as long-term persistence that are crucial in designing the turbines. In this work we aim to estimate the uncertainty in exerted forces using Hurst-Kolmogorov stochastic models and compare the results with those calculated using deterministic methods. A key objective is to examine how the fatigue of the structures is affected by the stochasticity of natural processes.

How to cite: Vrettou, S. E., Iliopoulou, T., and Koutsoyiannis, D.: Stochasticity of natural processes concerning loads exerted on offshore wind turbines, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-19137, https://doi.org/10.5194/egusphere-egu24-19137, 2024.

Flood hydrographs are desirable at various hotspots in river basins for a wide range of applications such as hydrologic design and risk assessment of water resources systems. However, determining the hydrographs at ungauged locations has always been challenging. Among various approaches available for simulating a flood hydrograph, the synthetic UH (unit hydrograph) approach has attracted the attention of hydrologists for use in ungauged catchments. The approach involves the derivation of UH for the target location’s catchment and convoluting it with an excess rainfall hyetograph specified for the catchment to arrive at a flood hydrograph. The UH can be derived using Geomorphological Instantaneous UH (GIUH). Recently, there has been growth in interest to consider Horton–Strahler (HS) ratio-based equivalent GIUH (E-GIUH) derived using the self-similarity hypothesis for ungauged catchments, owing to its advantages in overcoming uncertainty in HS ratios arising from uncertainty in the choice of a source DEM. The E-GIUH is constructed using a PDF (probability distribution function) that provides an adequate fit to salient points determined from the E-GIUH characteristics (peak flow, time to peak, and base time). The characteristics are derived using empirical relationships considering the catchment’s geomorphology (equivalent HS ratios, length of the highest order stream) and characteristic flow velocity. An issue in analysis with E-GIUH is that its construction involves uncertainty in the choice of PDF. Furthermore, the empirical relationships used with E-GIUH lack a physical basis. This paper proposes an entropy theory-based methodology to account for the uncertainty in the choice of a PDF for the E-GIUH construction. The efficacy of the proposed methodology in simulating flood hydrographs at ungauged sites is illustrated for typical rainfall-runoff events from 10 unregulated catchments ranging in size from 107 – 2000 km² and located in two river basins along the west coast of India. The hydrographs are compared with those simulated using conventional EGIUH in terms of several performance measures to illustrate the improvement noted with the entropy theory-based methodology.

How to cite: Goel, A. and Srinivas, V. V.: Entropy theory-based approach to derive equivalent GIUH, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1390, https://doi.org/10.5194/egusphere-egu24-1390, 2024.

North Euboea features elevated topography, expansive wild forests, and numerous small rivers with rainfall patterns similar to the Greek average. Since 2017 a severe disease has affected plane trees (Platanus orientalis), crucial for stabilizing water flow along riverbanks. A devasting wildfire in August 2021 (52,900 hectares burnt areas), further impacted the region significantly. These events induced substantial alterations to the landscape, changing the dynamic of waterflow in watersheds. Our research aims to comprehend, through statistical reasoning, the modifications in water flow and evaluate the collective repercussions on water management. For this purpose, systematic on-site inspections were conducted in April 2023 to map a substream/subriver of the Nileas River and to use water depth for the calibration of the 2D hydraulic model. Also, in September 2023, a devastating storm (“Elias”) caused a dramatic effect in riverbanksdue to the severity of the flood event, and so,in November 2023, we conducted a second systematic on-site inspection to comprehensively assess the situation after the flood event. Hence, a comparative study of Nileas River (before and after the storm) is presented in this study, by emphasizing, through statistical analyses, on the impacts of the landscape changes thoroughlyand to the understanding of the dynamic alterations in the region's watersheds.

How to cite: Moraiti, K., Sigourou, S., Kougia, M., Sargentis, G.-F., and Koutsoyiannis, D.: On-site inspection in monitoring of water flow for the calibration of hydraulic models and the statistical analysis of the hydraulic output. Case study: The river Nileas in North Euboea Greece, before and after the storm Elias in September 2023, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5186, https://doi.org/10.5194/egusphere-egu24-5186, 2024.

Groundwater discharge profoundly influences river flow, especially during dry spells, potentially exacerbating drought conditions, an issue compounded by escalating climate change-induced hydrological extremes. Amidst this, understanding aquifer systems' efficacy in mitigating (sub-)seasonal fluctuations and their ecological impacts gains significance even in moderate climates.

This work introduces a stochastic modeling approach for groundwater-fed baseflow, as an alternative to the traditional hydraulic theory by accounting for spatial heterogeneity of subsurface storage properties and associated uncertainties. Leveraging on readily available rainfall-generated recharge and river discharge series, stochastic tools determine baseflow characteristics. The model’s foundation lies on representing groundwater recharge and subsurface storage as stochastic variables.

With the subsurface storage represented as multiple linear reservoirs with stochastic storage parameters, the proposed model reveals the baseflow dynamics and the interplay between heterogeneous reservoir timescales and recharge variability, elucidating temporal variance of baseflow at the catchment scale. Furthermore, the study investigates equivalent parameters for an upscaled unique reservoir to model catchment behavior. Utilizing established stochastic analysis tools in subsurface hydrology, this research advances our understanding of heterogeneous hydrological catchments. In addition, the investigation analyzes the temporal statistical moments of baseline discharge dependent on input recharge and sub-catchments' response times. This detailed analysis unveils the system's attenuation effect as a metric for catchment resilience during prolonged droughts, significantly influenced by underlying hydrogeological properties. As a practical consequence, quantifying the dependency of the attenuation factor on both temporal and hydrogeological variability can help in identifying particularly sensitive watersheds, crucial for tailored adaptation strategies.

How to cite: Di Dato, M., Bellin, A., Cvetkovic, V., Dagan, G., Dietrich, P., Fiori, A., Teutsch, G., Zech, A., and Attinger, S.: A stochastic model of catchment baseflow dynamics, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2820, https://doi.org/10.5194/egusphere-egu24-2820, 2024.

Floods are a high-impact natural hazard in Switzerland and are responsible for considerable damage to property values, infrastructure and agricultural land (Hilker et al., 2009). Hence, reliable flood estimates are critical in flood risk reduction, emergency preparedness and disaster management.

The traditional flood estimation methods rely on statistical techniques based on observed streamflow and precipitation. However, the short length of observational records is often a limiting factor that leads to considerable uncertainties in flood estimates, especially for rare floods. An alternative approach that circumvents this limitation is the combination of stochastic weather generators with hydrological models using long continuous simulations. The advantage of this approach is that it avoids assumptions about antecedent catchment states (e.g., soil moisture, snowpack, storage levels of lakes and reservoirs) and simplified representations of the underlying physical flooding processes.

Here, we use an elaborate framework based on continuous simulations with a hydrometeorological modeling chain (Viviroli et al., 2022) to estimate rare floods for large catchments in Switzerland (larger than ~450 km²). The modeling chain starts with a multi-site stochastic weather generator (GWEX), focusing on generating extremely high precipitation events. Then, a bucket-type hydrological model (HBV) is used to simulate discharge time series. Finally, the RS Minerve (RSM) model is employed to implement simplified representations of river channel hydraulics and floodplain inundations.

We aim to investigate the uncertainties of derived flood estimates, with an experimental set-up focusing on the first component of the modeling chain, the weather generator. Aiming to explore the impact of precipitation inputs on flood estimates, GWEX is subject to different tests while the remaining components (HBV and RSM) remain unchanged. To achieve this, two experiments have been conducted: (a) parameterization of GWEX based on a bootstrap sampling of observed precipitation, from which we get an ensemble of 10 different synthetic time series (b) conditioning of the most relevant GWEX parameters to different weather types that describe intermediate, moderate and stronger precipitation intensities. A set of reference scenarios using the initial parameters of GWEX serves as a benchmark for comparison.

Our experimental framework unravels the sensitivity of the catchments to changes in precipitation inputs. While bootstrapping shows a higher impact compared to weather-type conditioning, the latter seems to reduce the spread of uncertainty both in precipitation and simulated floods. These findings provide an essential basis for follow-up studies on hazard assessment, safety analyses and hydraulic engineering projects.

References

Hilker, N., Badoux, A., & Hegg, C. (2009). The Swiss flood and landslide damage database 1972-2007. In Hazards Earth Syst. Sci (Vol. 9). www.nat-hazards-earth-syst-sci.net/9/913/2009/

Viviroli D, Sikorska-Senoner AE, Evin G, Staudinger M, Kauzlaric M, Chardon J, Favre AC, Hingray B, Nicolet G, Raynaud D, Seibert J, Weingartner R, Whealton C, 2022. Comprehensive space-time hydrometeorological simulations for estimating very rare floods at multiple sites in a large river basin. Natural Hazards and Earth System Sciences, 22(9), 2891–2920, doi:10.5194/nhess-22-2891-2022

How to cite: Kritidou, E., Kauzlaric, M., Vis, M., Staundiger, M., Seibert, J., and Viviroli, D.: Unraveling uncertainties of extreme-flood simulations over Switzerland based on different weather generator scenarios, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3346, https://doi.org/10.5194/egusphere-egu24-3346, 2024.

For optimal planning of reservoir design and management, long series or many realizations of daily streamflow are required. Stochastic streamflow models can provide such data. However, considering future changes in climate in these models is challenging. The objective of this study is to compare three non-stationary stochastic models with respect to their performance in simulating daily streamflow for current and future climate conditions. This comparison relies on two non-parametric approaches, namely the k-nn Bootstrap and Simulated Annealing optimization as well as a parametric model working in the frequency domain.

All models are run under different experiments: (1) with observed climate from the German Weather Service for a reference and pseudo-future period and (2) with future climate simulations using data from climate models. The simulations from the three models are evaluated for general flow statistics considering current climate and observed changes for a pseudo future. As an additional reference, the HBV rainfall - runoff model driven by observed climate and climate model data is used. The testing of the methods is carried out for some mesoscale catchments in the Harz Mountains comprising streamflow gauges with long daily records. The ability of the stochastic models to simulate the changes for the pseudo future will be the core test for their applicability under changing climate conditions. The results are also expected to demonstrate advantages, disadvantages and limitations of the three methods.

How to cite: Haberlandt, U., Jiang, Z., Brunner, M., Chartier-Rescan, C., Brandt, A., and Sharma, A.: Climate informed non-stationary simulation of daily streamflow – a comparison of three stochastic models, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10852, https://doi.org/10.5194/egusphere-egu24-10852, 2024.

We develop a regionalization framework for rainfall intensity-timescale-return period relationships that is implemented across the Greek territory. The methodology for single-site estimation is based on a stochastic framework for multi-scale rainfall intensity modeling. Five parameters are first independently fitted for each site, and the resulting parameter variability is explored in terms of uncertainty and spatial variability patterns. Two parameters, the tail-index and a timescale parameter, are identified as constant in space and estimated using data pooling techniques. The remaining three parameters are regionalized across Greece using a combination of spatial interpolation and smoothing techniques, which are evaluated using cross-validation in a multi-model framework.

How to cite: Iliopoulou, T., Koutsoyiannis, D., Malamos, N., Koukouvinos, A., Dimitriadis, P., Mamassis, N., Tepetidis, N., and Markantonis, D.: A stochastic framework for rainfall intensity-timescale-return period relationships regionalized over Greece , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9043, https://doi.org/10.5194/egusphere-egu24-9043, 2024.

The water, energy, and food nexus underscores the intricate interdependence among these vital resources, highlighting the need for integrated and rational management. Balancing the synergies and trade-offs within this nexus is crucial for addressing global challenges related to resource security, environmental management, and societal well-being. While the elements of the nexus are interdependent, various parameters hinder their causal connections. In this study we create a stochastic toy model which links each part of the nexus to the other based on Hurst-Kolmogorov dynamics. The introduction of simple environmental and social variables to these interdependencies alters the model dynamics of the nexus. In this way, we observe that despite the apparent causality in the dependencies of the nexus, the complexity of the system does not reveal causation. Therefore, as the nexus is vital for social prosperity, we must explore other methods for decision making.

How to cite: Markantonis, D., Matina, K., Koutsoyiannis, D., and Sargentis, G.-F.: Causality in Water-Energy and Food nexus: A toy-model multitasking stochastic synthesis, highlighting the complexity and randomness in decision making, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6102, https://doi.org/10.5194/egusphere-egu24-6102, 2024.

Successful modelling of the groundwater level variations in hydrogeological systems of complex formations considerably depends on spatial and temporal data availability and knowledge of the boundary conditions. Geostatistics plays an important role in model-related data analysis and preparation but has specific limitations when the aquifer system is inhomogeneous. In this research work, we show how the fusion of geostatistics with machine learning can solve some of these problems in complex aquifer systems, mainly when the available dataset is large and randomly distributed in the different aquifer types of the hydrogeological system. Self-Organizing Maps can be applied to identify locally similar input data, to substitute the usually uncertain correlation length of the variogram model that estimates the correlated neighborhood, and then by means of Transgaussian Kriging to estimate the bias-corrected spatial distribution of groundwater level. The proposed methodology was tested on a large dataset of groundwater level data in a complex hydrogeological district, and the results were significantly improved if compared to classical geostatistical approaches.

The research project is implemented in the framework of H.F.R.I call “Basic research Financing (Horizontal support of all Sciences)” under the National Recovery and Resilience Plan “Greece 2.0” funded by the European Union – NextGenerationEU (H.F.R.I. Project Number: 16537).

How to cite: Varouchakis, E.: Fusion of geostatistics and machine learning under a stochastic approach for the spatial analysis of groundwater level variations , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10826, https://doi.org/10.5194/egusphere-egu24-10826, 2024.

Aquitards are important hydrogeological features and their properties play an important role in e.g. water resources management, subsidence, aquifer thermal energy storage and contamination transport. The hydraulic conductivity of aquitards is typically parameterized by analytical interpretation of pumping tests or model calibration. However, these methods may not be very accurate for aquitards and usually do not account for spatial variability and uncertainty. Alternatively, core-scale measurements of hydraulic conductivity are used in geostatistical upscaling methods, for which their correlation lengths are needed. However, this information is extremely difficult to obtain. In this study we investigate whether readily available data from three drinking water extraction sites in The Netherlands can be used to obtain the geostatistical parameters of aquitard hydraulic conductivity needed for upscaling and to provide information about spatial variability of the hydraulic conductivity. We generated conditional random realizations from core scale data with varying correlation lengths and lithology distributions, upscaled these to model block scale and inserted these into a groundwater flow model that simulates the impacts of drinking water extraction and natural variability in precipitation and evapotranspiration over an extensive time period. We select the realizations that best fit observed groundwater heads to extract information about aquitard correlation lengths and lithology distributions and derive upscaled spatially varying aquitard conductivities.

How to cite: van Leer, M., Zaadnoordijk, W. J., Zech, A., Griffioen, J., and Bierkens, M.: Investigating aquitard heterogeneity by inverse groundwater modelling of drinking water extraction sites, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9090, https://doi.org/10.5194/egusphere-egu24-9090, 2024.

This study outlines a systematic workflow for developing a 3D geological model with the goal of converting it into a groundwater flow model to simulate groundwater dynamics in the lower Spree catchment in Berlin/Brandenburg, Germany. The flow model is constructed by employing  the finite-difference method to discretize groundwater flow equations, focusing on investigating saltwater up-coning resulting from changes in recharge or increased pumping.

For this aim geological cross-section profiles and borehole data are used to construct a 3D geological model covering 1100 km² with depths up to 350 m. This model serves as a solid base for the subsequent 3D flow model with fine spatial resolution (100 m horizontally and 5 m vertically) and monthly temporal resolution.

The foundation of the geological model hinges on the stratigraphic order and addresses structural complexities such as faults, folds, dip angles, azimuth, and extensions. Challenges arise from outdated data, requiring meticulous preprocessing and cleaning, especially for larger areas with intricate structures and lenses. Specific challenges involve multiple boreholes sharing identical coordinates with varying lithological and stratigraphical descriptions for the same depth, as well as the repetition of layers. Addressing these issues requires careful preprocessing and cleaning.

The study explores interpolation techniques, including Inverse Distance Weighting and Natural Neighbor algorithms, commonly used in geospatial analysis. Challenges related to these methods are discussed, emphasizing difficulties in dealing with shallow boreholes, missing stratigraphy, and ensuring layer continuity, affecting the overall reliability of interpolation results.

To overcome these challenges, a Multi-Layer Perceptron (MLP) machine learning classifier is introduced. This classifier learns the hierarchical order of lithological information, seamlessly integrating it into the geological model. The MLP classifier is trained on preprocessed data, utilizes a dataset split into 85% for model training and 15% for validation, achieving a validation score of 73%.

The geological model is then converted into a numerical mesh for the groundwater flow model. Hydraulic parameters, including hydraulic conductivity, porosity, specific storage, and specific yield, are estimated using empirical formulas and correlation sheets such as the Hazen and Bayer methods, which involve the determination of effective grain size (D10 and D30). Statistical algorithms aid in identifying and assigning hydraulic parameters related to the dominant material group to each flow cell. In instances where two material groups exhibit the same dominance, average values of hydraulic parameters are calculated.

In conclusion, the study highlights the utilization of advanced techniques, including machine learning, alongside statistical methods, becomes imperative to solve complex geological settings, ensuring a more accurate and reliable representation of subsurface properties for groundwater flow models.

How to cite: Abdelrahman, A. A. A., Engelhardt, I., and Sauter, M.: Integration of complex geological systems into groundwater modelling to simulate saltwater dynamics’, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11894, https://doi.org/10.5194/egusphere-egu24-11894, 2024.

The designation of nitrate-polluted areas for groundwater protection based on national directives implementing the EU Nitrates Directive (91/676/EEC) in Germany requires the use of geostatistical or deterministic regionalization methods. The objective of this study was to assess the applicability and propose a suitable methodology for a possible national-scale area designation based on an in-depth problem analysis and empirical as well as theoretical model assessments, which identified shortcomings, uncertainties and possible biases in the methods used until now.

Suitable methods must not only be able to identify exceedance regions – as opposed to simply regionalizing nitrate concentration; they also need to take into account spatial heterogeneity and adequately represent distributional characteristics across a variety of hydrogeological settings. This requires the incorporation of ancillary information in the form of quantitative and categorical spatial predictors representing hydrogeological and general environmental conditions, but not emissions estimates in the present regulatory setting.

Kriging with external drift, geostatistical regression-kriging methods and conditional geostatistical simulations offer an established methodological toolbox that fulfils these requirements and enables transparent decision-making. These models consist of linear, potentially nonlinear spatial trend components and geostatistical interpolation components, which can be further differentiated based on hydrogeological regions to account for heterogeneity. An unbiased estimate of the total exceedance area with nitrate levels >50 mg/l can be obtained from these models and accounted for in the decision-making process. An empirical comparison highlights possible biases in the size of exceedance areas obtained with traditional approaches that ignore local prediction uncertainty and focus on spatial prediction of nitrate concentration. Potentials and challenges of combining geostatistical techniques with nonlinear machine-learning models in a regulatory context are discussed.

How to cite: Brenning, A. and Suesse, T.: Regionalisation methods for the designation of areas with groundwater nitrate pollution in Germany, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15800, https://doi.org/10.5194/egusphere-egu24-15800, 2024.

Hydrology evaluates habitat diversity and quality by identifying where living organisms prefer the environment through habitat information. This information significantly influences biodiversity conservation and the survival of various organisms. Moreover, understanding the types and characteristics of aquatic organisms contributes to determining water quality. This study used a hydrological approach to understand the interaction between the environment and organisms with data on the geographical distribution of freshwater organisms and the various water quality conditions in their respective habitats. Specifically, fuzzy coding was applied to integrate the geographical distribution of over 200 macroinvertebrates in Korea with various water quality conditions (i.e., pH, total phosphates, total nitrogen, etc.) in their preferred habitats. Fuzzy coding quantified the affinity of species for various water quality parameters (e.g., pH) composed of ambiguous or overlapping modalities (e.g., acidic, neutral, and base), constructing fuzzy-coded species x environmental gradient matrices for each species. The resulting database will be critical in evaluating habitat diversity and quality for aquatic organisms from a hydrological perspective. It can be used as an indicator for habitat and biodiversity conservation. It will also provide important information for assessing organisms' adaptability to environmental changes based on their preferences for water quality conditions. Moreover, the database can contribute to establishing policies on water pollution and conservation by understanding the water characteristics preferred by a given organism. Further study is needed to understand the ecological impact of habitat changes due to environmental variations and appropriate modeling to predict water quality changes.

How to cite: Park, J., Yoon, J.-H., and Kim, S. D.: Evaluation of habitat diversity and water quality preferences of macroinvertebrates through fuzzy coding, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8121, https://doi.org/10.5194/egusphere-egu24-8121, 2024.

The biogeochemistry of wetland ecosystems is driven by the presence and absence of water. Wetlands are known hotspots of methane (CH₄) emissions, particularly when inundated. Monitoring short-term, and possibly small-scale changes in inundation is therefore critical to quantifying both local and global CH₄ emissions. Despite their importance, these short-term changes have historically been under-reported in efforts to monitor CH₄. As sea levels rise and flood events increase, it’s imperative to account for these events to better project CH₄ cycle variation in a changing climate. Remote sensing is the only method capable of monitoring these changes over time at scale; however, no current remote sensing product has the spatial and temporal resolutions required to map ephemeral changes in inundation extents accurately. To address this, we developed a method to generate high spatiotemporal resolution inundation maps combining SAR and optical data from Sentinel-1 and Sentinel-2 imagery supplemented with commercial PlanetScope imagery from 2017–2022. This method was evaluated in the Albemarle-Pamlico Peninsula, a coastal wetland region in North Carolina, United States characterized by frequent and variable inundation.

Two decision-tree based machine learning algorithms were tested to map inundation extents: a random forest (RF) model and an extreme gradient boosted (XGBoost) model. The models were trained for each sensor based on a suite of spectral signals, terrain-derived features, and precipitation data for each image at the sensor’s native resolution. This work revealed minor differences between machine learning classifiers across the 5 years, with RF accuracies of 94.0%, 98.2%, and 98.6% and XGBoost accuracies of 89.1%, 98.3%, and 97.8% for PlanetScope, Sentinel-2, and Sentinel-1 respectively. The RF classified inundation maps from each sensor were then fused using a hierarchical spatiotemporal random effects model within a probit link function, to generate daily time series of inundation probabilities at 5 m resolution. This approach is unique in that we 1) address the differing sensor resolutions using a statistical change-of-support formulation with observations mapped to process locations, 2) fuse non-Gaussian (binary) responses from machine learning outputs, and 3) model spatial and temporal autocorrelation through spatial basis functions and a first-order autoregressive time series model. Overall, this work produced a novel 5-year inundation dataset, capturing both long-term and ephemeral changes in inundation extents that are critical for quantifying components of the water cycle and their interactions with biogeochemical cycles on Earth.

How to cite: Abrahamson, J., Gray, J., and Schliep, E.: Multi-Sensor Space-Time Data Fusion of Machine Learning Generated Time Series for Wetland Inundation Monitoring, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11476, https://doi.org/10.5194/egusphere-egu24-11476, 2024.