Session assets
Orals: Tue, 29 Apr | Room -2.92
Large-scale atmospheric vortices like the polar vortex or the Asian monsoon anticyclone are known to confine compounds for several months in the corresponding regions of the stratosphere with many consequences on the transport and the resulting atmospheric composition, the chemical activity and radiative properties.
It was recently discovered that confinement over the same time scale occurs also in much smaller mesoscale anticyclonic vortices generated within the absorbing plumes of smoke or ash generated by large forest fires and some volcanic eruptions.
As a rule, the atmosphere dissipates rapidly all inertial structures and these vortices are all maintained by a sustained forcing. We will discuss the similarities and differences among those vortices, the smoke vortices being distinguished by their autonomy as they carry their own source of forcing when they travel around the globe.
We will discuss the phenomenon of isentropic vortex shedding which is a main mechanical dissipation factor and show that it behaves very similarly at all scales. In the vertical direction, the flux processor of the large vortices will be compared to and distinguished from the leaking process of the rising smoke vortices. Other processes associated with radiative relaxation of thermal anomalies play role both to maintain and dissipate.
Although the state of understanding of smoke vortices is still very incomplete, a discussion of their condition of formation, maintenance and stability will be offered based on observations and idealized numerical simulation.
How to cite: Legras, B., Podglajen, A., Rezig, M., and Duchamp, C.: Confinement and shedding, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-13384, https://doi.org/10.5194/egusphere-egu25-13384, 2025.
How to cite: Bhowmick, T., Falkinhoff, F., Bodenschatz, E., and Bagheri, G.: Lagrangian tracking of particles settling through the atmosphere: influence of particle shape on its dispersion, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3171, https://doi.org/10.5194/egusphere-egu25-3171, 2025.
Turbulent relative dispersion is a phenomenon of fundamental interest both for its theoretical implications and for its immediate applications, which in geophysical sciences range from pollutant spreading in the atmosphere to nutrient transport in the oceans. We present new results in the measurement of turbulent relative dispersion in the atmospheric boundary layer, which enrich the picture with respect to the current framework. The measurements were carried out using clusters of miniaturized radiosondes, carried by small (~40 cm in diameter) helium balloons [1]. These clusters enable the effective investigation of relative dispersion by computing inter-particle distances among radiosondes. This methodology represents a concrete attempt at realising the type of analysis originally conceived by L. F. Richardson in his 1926 paper [2], often regarded as the one that initiated the field of study of relative dispersion.
The current accepted framework for the discussion of relative dispersion is the Kolmogorov-Obukhov scaling theory, which on dimensional grounds allowed for the derivation of the result (called the Richardson-Obukhov law) according to which the mean square distance in between particles advected by a turbulent flow field scales like the cube of time, , where ε is the energy dissipation rate and g is called the Richardson constant. However, this result is only valid for the case of homogeneous, isotropic turbulence, specifically in the inertial range of scales [2, 3]. Atmospheric turbulence, instead, is far from homogeneity and isotropy, and is characterized by local intense intermittency and entrainment [4, 5].
We conducted six cluster launches across three distinct topographical environments: the near-maritime plains at Chilbolton Observatory, the western Alps near the Astronomical Observatory of Aosta Valley, and the hilly region surrounding Udine. The results reveal not only deviations from the RO law but also significant variations between launches and distinct dispersion regimes within each launch (Fig. 1). The implication is, as expected, that the dispersion law for the atmosphere does not have a universal character, and depends on specific details of the boundary layer flow. The next step in the analysis will be the identification of the relevant flow features which impact the dispersion law, which is especially challenging due to the wide range of possibly participating phenomena.
Fig. 1. Mean square separation distance between mini-radiosondes within the cluster during six field experiment flights in different environmental topologies. Cross symbols show results from the MET OFFICE NAME dispersion model [6].
1. Abdunabiev S. et al., Measurement 224, 113879 (2024)
2. Richardson LF, Proc. R. Soc. Lond. A 110, 709 (1926)
3. Devenish, BJ, Thomson DJ. JFM 867, 877–905 (2019)
4. Van Reeuwijk M, Vassilicos JC, Craske J. JFM 908 (2021)
5. Fossa’ L., Abdunabiev S., Golshan M., Tordella D., Physics of fluids 34, (2022)
6. Turbulence_&_Diffusion_Note_288, Met Office, UK (2003)
How to cite: Gallino, N., Abdunabiev, S., Craske, J., Devenish, B., Tse, J., and Tordella, D.: New ABL measurements of Lagrangian relative dispersion by means of radiosonde clusters, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-6918, https://doi.org/10.5194/egusphere-egu25-6918, 2025.
Ocean flows at scales smaller than few hundreds of kilometers display rich dynamics, mainly associated with quasi- geostrophic motions and internal gravity waves. Although both of these processes act on comparable lengthscales, the former, which include meso and submesoscale turbulent flows, are considerably slower than the latter, which take part in the ocean fast variability. Understanding how their effects overlap is crucial for several fundamental and applied questions, including the interpretation and exploitation of new, high-resolution satellite altimetry data, and the characterization of material transport at fine scales.
In this study we investigate these points by examining Lagrangian pair-dispersion statistics in a high-resolution global-ocean numerical simulation including high-frequency motions, such as internal gravity waves. In particular, we aim at assessing the sensitivity of the particle relative-dispersion process on ageostrophic, fast fluid motions. For this purpose we select a study area close to Kuroshio Extension, characterized by energetic submesoscales, and focus on the seasonal variability of the Lagrangian dynamics.
We find that in winter pair dispersion is predominantly influenced by meso and submesoscale motions, meaning nearly balanced dynamics. The behavior of the different Lagrangian indicators considered agrees in this case with the theoretical predictions, based on the shape of the kinetic energy spectrum, in quasi-geostrophic turbulent flows. Conversely, in summer, when high-frequency motions gain importance and submesoscales are less energetic, the situation is found to be more subtle, and the usual relations between dispersion properties and spectra do not seem to hold. We explain this apparent inconsistency relying on a decomposition of the flow into nearly-balanced motions and internal gravity waves. Through this approach, we show that while the latter contribute to the kinetic energy spectrum at small scales, they do not impact relative dispersion, which is essentially controlled by the nearly-balanced, mainly rotational, flow component at larger scales.
How to cite: Berti, S., Maalouly, M., Lapeyre, G., and Ponte, A.: Relative dispersion at the surface of the ocean: role of balanced motions and internal waves, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5876, https://doi.org/10.5194/egusphere-egu25-5876, 2025.
The Southern Ocean plays a crucial role in regulating Earth's climate, absorbing approximately 10 % of annual human CO2 emissions (DeVries, 2014; Friedlingstein et al., 2023). However, it is still a challenge to fully understand its carbon dynamics due to significant observational gaps, particularly during winter. Furthermore, the dynamics on small spatial and temporal scales are also poorly understood, despite their potential importance in shaping the overall carbon budget of the region (Guo & Timmermans, 2024). Between 2001 and 2012, researchers from the LOCEAN laboratory in Paris deployed 15 CARIOCA Lagrangian drifting buoys in this region to gain a deeper understanding of its spatial carbon dynamics (Boutin et al., 2008; Resplandy et al., 2014) at high-frequency (1-hour time resolution). In this study, we analyzed the time series of six of these buoys in the framework of Lagrangian turbulence (Kolmogorov, 1941; Landau & Lifschitz, 1944; Inoue, 1951). This is done using Lagrangian data on CO2 fugacity (fCO2), chlorophyll a, sea surface temperature (SST), and sea surface salinity (SSS) along their trajectories. Additionally, we examined the dynamics of the buoys' drifting speeds estimated from buoys location data.
Through Fourier spectral analysis and structure function analysis, we demonstrated that these time series exhibit scaling and intermittent behaviour, in agreement with the Lagrangian vision of the turbulent Richardson-Kolmogorov energy cascade in fully developed turbulence. Notably, at least two distinct spectral regimes were identified. Chlorophyll a and fCO2 behave as active turbulent scalars, while SST and SSS depicted statistics compatible with passive scalars with a higher intermittency on timescales shorter than 4 days. The links between these time series were also investigated, using the generalized correlation functions (GCFs) and exponents (GCEs).
References :
DeVries, T. (2014). The oceanic anthropogenic CO2 sink: Storage, air‐sea fluxes, and transports over the industrial era. Global Biogeochemical Cycles, 28(7), 631-647.
Friedlingstein, P. et al. (2023). Global carbon budget 2023. Earth System Science Data, 15(12), 5301-5369.
Guo, Y., & Timmermans, M. L. (2024). The role of ocean mesoscale variability in air‐sea CO2 exchange: A global perspective. Geophysical Research Letters, 51(10), e2024GL108373.
Boutin, J. et al. (2008). Air‐sea CO2 flux variability in frontal regions of the Southern Ocean from CARbon Interface OCean Atmosphere drifters. Limnology and Oceanography, 53(5part2), 2062-2079.
Resplandy, L. et al. (2014). Observed small spatial scale and seasonal variability of the CO2 system in the Southern Ocean. Biogeosciences, 11(1), 75-90.
Kolmogorov, A. N. (1941). On degeneration (decay) of isotropic turbulence in an incompressible viscous liquid. In Dokl. Akad. Nauk SSSR (Vol. 31, pp. 538-540).
Landau L.D. & Lifshitz E.M. (1944). Fluid Mechanics (MIR), first russian edition.
Inoue, E. (1952). Turbulent fluctuations in temperature in the atmosphere and oceans. Journal of the Meteorological Society of Japan. Ser. II, 30(9), 289-295.
How to cite: Robache, K. and Schmitt, F. G.: Turbulent Lagrangian fCO2 time series statistics in the Southern Ocean, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-473, https://doi.org/10.5194/egusphere-egu25-473, 2025.
Mesoscale eddies play a pivotal role in oceanic dynamics, influencing transport, mixing, and energy distribution. Current detection methods are primarily divided into Eulerian and Lagrangian approaches, each highlighting unique eddy characteristics. Eulerian methods rely on instantaneous fields, such as sea surface height, Okubo–Weiss parameter or vorticity, to identify the eddy boundaries. In contrast, Lagrangian approaches utilize water parcel trajectories to compute metrics like the Lagrangian Average Vorticity Deviation (LAVD) or Finite-Time Lyapunov Exponents (FTLE), identifying rotationally coherent Lagrangian vortices (RCLVs) with minimal exchange across the boundary. Eulerian eddies, however, are inherently "leaky", allowing for fluid exchange due to the fact that their boundaries are non-material. Despite these differences, both approaches capture complementary aspects of the same physical phenomenon. This study aims to bridge the gap between the two eddy detection methods by combining their strengths and leveraging high-resolution simulations from the coupled climate model ICON. Here, we identify daily RCLVs from evolving LAVD fields to find the time at which each Eulerian eddy loses coherence. In doing so, we will be able to explore how eddy coherence changes though its lifecycle and geographical location. This combined methodology can deepen our understanding of mesoscale ocean transport by quantifying realistic eddy trapping ability.
How to cite: Bērziņa, S., Wienkers, A., Gruber, N., and Münnich, M.: On uniting Eulerian and Lagrangian mesoscale eddy perspectives, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-14793, https://doi.org/10.5194/egusphere-egu25-14793, 2025.
In past decades, increasing robust causal models were proposed, making causal inference under different scenarios and data limitations feasible. On one hand, these causal model are all based on time series data sources. On the other hand, in Earth Science, some variables, such as soil features and elevation, do not present a time series or the time series of these variables do not present sufficient temporal variations. In this case, traditional temporal causal models may fail to identify these clearly existing causalities in Earth Science. To fill these gaps, here we show a Geographical Convergent Cross Mapping (GCCM) model for spatial causal inference with spatial cross-sectional data based cross-mapping prediction in reconstructed state space. Three typical cases, where clearly existing causations cannot be measured through temporal models, demonstrate that GCCM could detect weak-moderate causations when the correlation is not significant. And when the coupling between two variables is significant and strong, GCCM is advantageous in identifying the primary causation direction and better revealing the bidirectional asymmetric causation, overcoming the mirroring effect. The principle and some cases of GCCM are briefly introduced.
How to cite: Chen, Z.: Causal Inference in GeoScience: From the Temporal to Spatial Dimensions, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-7326, https://doi.org/10.5194/egusphere-egu25-7326, 2025.
How to cite: Cummins, D. P. and Li, M.: Sparse pre-whitening operators for regression of climatic time series, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11509, https://doi.org/10.5194/egusphere-egu25-11509, 2025.
Nonlinear quadratic and linear dynamical dependencies of large-scale climate modes are disentangled through the analysis of the rate of the information transfer. That is performed in a joint analysis of eight dominant climate modes, covering the tropics and extratropics over the North Pacific and Atlantic. A clear signature of nonlinear and compound influences at low-frequencies (time scales larger than a year) are emerging, while high-frequencies are only affected by linear dependencies. These results point to the complex nonlinear collective behavior at global scale of the climate system at low-frequencies, supporting earlier views that regional climate modes are local expressions of a global intricate low-frequency variability dynamics, which is still to be fully uncovered.
How to cite: Vannitsem, S., Liang, X. S., and Pires, C. A.: Nonlinear causal dependencies as a signature of the complexity of the climate dynamics, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-4715, https://doi.org/10.5194/egusphere-egu25-4715, 2025.
The Mediterranean Sea is characterized by an anti-estuarine circulation, with Atlantic Water entering the Strait of Gibraltar at the surface and denser waters, formed within the basin, exiting at depth as the Mediterranean Outflow. Early studies identified the Western Mediterranean Deep Water, formed in the Gulf of Lions, as the primary source of the dense water masses contributing to the Outflow. While confirming this finding, more recent analyses of in-situ observations have highlighted additional contributions from other intermediate and deep water masses, such as Western Intermediate Water, Levantine Intermediate Water and Tyrrhenian Deep and Intermediate Waters.
In this study, the origin of the Mediterranean Outflow is investigated by deploying six million Lagrangian parcels at the Strait of Gibraltar, and advecting them backward in time using velocity estimates from an eddy-permitting reanalysis. Trajectories are integrated until parcels reach one of three origin sections within a maximum time of 78 years. To estimate the transport exchange between the origin sections and the Strait of Gibraltar, each parcel is tagged with a small volume transport, which is conserved along the trajectories due to the non-divergence of the velocity field.
The results indicate that 86% of the Outflow's transport originates from the Gulf of Lions, associated with Western Mediterranean Deep Water and Western Intermediate Water; 13% from the Strait of Sicily, related to Levantine Intermediate Water; and 1% from the Northern Tyrrhenian, related to Tyrrhenian Deep and Intermediate Waters. Mediterranean dense waters all recirculate in the Algerian Basin and in the deep Tyrrhenian basin, where stirring and mixing processes are hypothesized to occur. Before exiting the Strait of Gibraltar, anticyclonic recirculation induced by the western Alboran gyre decreases the density and depth of the water mass, ultimately shaping the characteristics of the Mediterranean Outflow. Temperature-salinity histograms at each origin section exhibit broad distribution, with peaks corresponding to expected water-mass types. The median transit times from the sections to the Strait of Gibraltar range from 5 years (Gulf of Lions) to 8 years (Strait of Sicily).
How to cite: Vecchioni, G., Cessi, P., Pinardi, N., Rousselet, L., and Trotta, F.: A Lagrangian Estimate of the Mediterranean Outflow's Origin, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-19212, https://doi.org/10.5194/egusphere-egu25-19212, 2025.
The Biomass Burning Aerosol Campaign (BiBAC) was conducted in the Kruger National Park (KNP), at Skukuza in South Africa during the 2022 biomass burning season. The campaign included an Intensive Observation Phase (IOP) from September to October, aiming to quantify aerosol optical properties and plume transport.(Ranaivombola et al., 2024). The combination of ground-based (sun-photometer), satellite observations (MODIS, IASI and CALIOP), and CAMS reanalysis show a significant aerosol and carbon monoxide (CO) loading linked to biomass burning activity. Using AOD data from sun-photometer observations, Ranaivombola et al., (2024) define two events of biomass burning plume over the Southwestern Indian Ocean (SWIO) basin: September 18 to 23 and October 9 to 17, called here after event 1 and event 2, respectively.
During Event 1, the plume was transported toward the SWIO basin as a "river of smoke" phenomenon. As reported previously in the literature (Swap et al., 2003 and Flamant et al., 2022), the meteorological conditions were influenced by the passage of westerly waves associated with a cut-off low (COL) that favored the eastern transport pathway. However, it was not the case during Event 1. There were two troughs which supported the formation of two frontal systems and contributed to the transport of aerosols and CO plumes from South America (SAm) towards Southern Africa (SA). This transport was driven by a westerly baroclinic wave through the mid-tropospheric layers.
Event 2 involved a more complex synoptic setup with three frontal systems supported by three distinct troughs, allowing the recirculation of plumes over SA. This dynamic system enhanced the transport of CO plumes from SAm, which merged with African plumes over the Mozambique Channel. The sustained activity of the baroclinic wave generated new troughs, keeping aerosol levels high for an extended period of 1.5 week. The progression of baroclinic waves and frontal system development were essential in driving regional and intercontinental transport of aerosols and CO plumes.
These two events allowed to reveal two transport mechanisms of aerosol plumes and CO between SAm and SA towards the SWIO basin. It shows also that SA is a target region for aerosols and CO from SAm biomass burning. To assess and quantify the contributions of SA and SAm sources to observed CO concentrations over SA, we used the FLEXPART model (version 10.4) coupled with CO emissions database (biomass burning and anthropogenic emission from CAMS: GFAS and CAMS-GLOB-ANT, respectively). Each simulation tracked particles representing CO back in time over a period of 20 days, during the IOP. The setup included daily releases of 20,000 particles over six sites in Southern Africa (Skukuza, Durban, Maun, Upington, Mongu and Gobabeb). Both SA and SAm sources significantly influenced the CO balance over SA. The contribution of biomass burning emissions from SA were higher than those from SAm. Nevertheless, the biomass burning emission from SAm were more variable and could occasionally match or exceed those from SA. This quantification confirmed the predominance of African sources but also highlighted the presence of intercontinental transport which is poorly investigated until now.
How to cite: Ranaivombola, M., Bègue, N., Krysztofiak, G., Vaz Peres, L., Sivakumar, V., Berthet, G., Jegou, F., Piketh, S., and Bencherif, H.: An Intensive Biomass Burning Aerosol Observation phase in 2022, over Skukuza, South Africa: CO transport and balance over Southern Africa, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17867, https://doi.org/10.5194/egusphere-egu25-17867, 2025.
Particles transport can be used to study flow fields on different scales in geophysics. Tracer and inertial particle transport can highlight the connectivity between different locations in the ocean or describe changes in flow fields based on the visualization of the flow by tracers. Of particular interest are long-range transport properties to either identify changes in flow paths due to climate change or to study the transport of seeds over long distances to identify sources of plants in different parts of the world. Such studies require particle tracking algorithms which are capable to work properly on a global scale of the Earth, i.e. on a spherical geometry.
We have created a sophisticated software tool that simulates the movement of large numbers of tracer and inertial particles within interpolated oceanic velocity fields, in our examples based on the publicly available HYCOM data. Built in C++ and parallelized with Intel Threading Building Blocks (Intel TBB), it achieves high performance when dealing with substantial computational loads. To accelerate nearest-neighbor searches, we organize the grid points into a kd-tree, making it quick to locate grid points near any particle. We then interpolate the eastward and northward velocity components using a Gaussian-shaped weight function — an effective choice that avoids the singularities sometimes encountered in inverse distance interpolation. Since planar projections can introduce significant distortions on a global scale, we also account for Earth’s spherical geometry. Specifically, we solve two-dimensional tracer equations and the Maxey–Riley equation for inertial particles on a local tangent plane. Afterward, we revert the computed particle positions to latitude-longitude coordinates via an azimuthal equidistant projection, mitigating large-scale errors in simulations that may span thousands of kilometers.
The software is capable of simulating the dispersal of seeds and algae by ocean currents, easily managing hundreds of thousands of particles under varied initial conditions. It reconstructs connectivity maps between distant coasts, identifies transport barriers through finite-time Lyapunov exponent calculations, and can compute derivatives of the velocity field — such as divergence, vorticity, and the Okubo–Weiss parameter — broadening its range of oceanographic applications.
We highlight the software’s capabilities with two representative examples. First, we track the origins of particles (such as plant seeds) and explore their possible routes to Hawaii. Second, we assess the likelihood that harmful algal blooms could drift into the Baffin Bay during the warmest parts of the summer.
How to cite: Kruglov, V. and Feudel, U.: Numerical studies of connectivity and Lagrangian transport in the world’s oceans, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11897, https://doi.org/10.5194/egusphere-egu25-11897, 2025.
This work presents experimental results regarding rotating penetrative convection. The focus is on how convection driven by thermal or salty composition interacts with a stably stratified region. In such systems, as convection overshoots into the stratified layer, complex feedback loops arise, leading to the generation of internal waves that propagate in the stably stratified region. In rotating systems, Coriolis effects can further modify the dynamics, giving rise to inertial-gravity waves in the stably stratified region. Furthermore, the convective cells can change into different patterns of elongated vortices, changing how convection overshoots, and how it can drive internal waves. These phenomena are relevant to different geophysical and astrophysical applications, such as in the Earth's atmosphere, where internal gravity waves are excited in the stratosphere by convective motions in the troposphere. These interactions are also relevant to planetary and stellar interior applications, where convection can drive waves in stably stratified layers such as the radiative zone of stars or in the (possibly existing) stratified layer at the Earth's external core, where rotation effects are even more significant due to the small Rossby numbers, of the order of $10^{-5}$ to $10^{-4}$. This indicates that rotational forces dominate over inertial forces, highlighting the importance of better understanding the effects of rotation in the dynamics of penetrative convection and wave interactions.
Our experimental setup, named \textit{CROISSANTS (Convective ROtational Interactions with Stable Stratification Arising Naturally in Thermal Systems)}, found at the Physics Laboratory of the Ecole Normale Supérieure (ENS) de Lyon, is mounted on a rotating table and investigates the dynamics of rotating systems using water with a temperature gradient. The temperature ranges from approximately $30^oC$ at the top of a $30$cm-high cubic cavity and decreases to $0^oC$ at the bottom. Since water exhibits a density inversion between $0^oC$ and $4^oC$, the system naturally develops convection at the bottom, beneath a stably stratified region that extends from the convective interface to the top of the cavity. Measurements were performed using techniques such as Particle Image Velocimetry (PIV), Schlieren techniques, and Laser-Induced Fluorescence (LIF), to capture the convective and wave motions in both vertical and horizontal planes. Numerical simulations complement the experiments, exhibiting similar behavior to the observed experimental results. Both experiments and numerical simulations show that the elongated vortices in the convective region can be observed in higher regions of the stable density stratified zone. These long-lasting vortices move slowly in the flow (compared to the rotation of the experiment). Lagrangian-Averaged-Vorticity-Deviation (LAVD) techniques are then applied to track the dynamics of these long-lasting vortices elongated in the stable region. Understanding these processes provides a framework for interpreting how convective motion transfers energy across scales, impacting large-scale magnetic fields and planetary evolution.
How to cite: Meletti, G., Alboussière, T., Curbelo, J., Labrosse, S., and Odier, P.: Vortex dynamics on Rotating Penetrative Convection, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11547, https://doi.org/10.5194/egusphere-egu25-11547, 2025.
The Orust-Tjörn fjord system is located on the west coast of Sweden and consists of several fjords connected by shallow and narrow straits. It is home to nature reserves, harbors, leisure areas, and aquaculture farms, and biodiversity is threatened by invasive species, for example brought in through shipping. Therefore, it is important to understand how larvae of invasive species are dispersed by the currents within the fjords system in order to find efficient methods for management of existing and future harmful invasive species.
A connectivity study is performed in order to identify dispersion patterns, assuming that larvae are passively transported by surface currents. For the years 2016 and 2022, the dispersion of larvae is simulated using the open source Opendrift software (Dagestad et al., 2018). The model is forced by velocity fields modeled with a high resolution regional Nemo3.6 ocean model with lateral resolution of ~50m. A large number of particles (~700,000) are seeded with four-day intervals, covering the whole fjord system and areas of open waters near the entrances of the fjord system. For each seeding, the dispersion model runs for 3 weeks where larvae that reach a shore are transported away again when currents change (pelagic phase). This is followed by a one-week period where larvae settled as soon as they reach a shore (settling phase). On the basis of this ensemble, we perform a connectivity analysis indicating the probabilities of larvae, released at one location, settling in other locations within the fjord system.
Results show that larvae seeded inside the Orust-Tjörn fjord system mostly remain there with some even remaining in the same local fjord basin during the four-week period. Connectivity matrices also show that some larvae travel far. Larvae seeded outside the Orust-Tjörn fjord system are likely to leave the model domain as they are transported within the generally northward flowing swift Baltic current.
How to cite: Brunnabend, S.-E., Arneborg, L., and Fredriksson, S.: Dispersal of invasive species larvae within the Orust-Tjörn fjord system, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-1290, https://doi.org/10.5194/egusphere-egu25-1290, 2025.
Well log curves, acquired from downhole logging tools during well logging, are pivotal for reservoir characterization and formation evaluation in oil and gas exploration and production. However, manual feature extraction from raw curves remains essential for constructing effective machine learning models, presenting time-consuming challenges and stringent labeling requirements. Concurrently, the transformer architecture, prevalent in NLP and computer vision, offers promise for representation learning. This paper proposes a self-supervised transformer based methodology for extracting well log curves representations, aiming to expedite downstream model development.
While transformer models have gained prominence in handling text and image data, well log curves present a distinct challenge as they resemble time series data. Despite the nascent development of time series transformer models, we conducted an extensive review of current progress and adopted the best-performing time series transformer model for extracting representations from well log curves. Importantly, given the challenges posed by factors such as borehole conditions and instrument failure, certain types of well log curves may occasionally be missing or distorted. To address this issue, our proposed methodology introduces an adaptive masking mechanism, which selectively applies masking to patches of curves where data quality is poor, thereby effectively mitigating data quality concerns.
Data from 2000 wells are utilized for model training, with an additional 100 wells reserved for validation purposes. Our study observed a consistent decrease in both training and test losses until convergence during the training stage. Initially, mean squared error (MSE) and mean absolute error (MAE) are employed to quantify reconstruction errors between reconstructed curves and raw curves, low values of MSE (0.08) and MAE (0.07) indicate effectiveness of the learned representations. Subsequently, a downstream task involving oil and gas identification is undertaken, wherein a classification model is developed based on representations learned by the transformer model. Performance comparison between models utilizing learned representation and those employing statistical features highlights the superior performance of the former (98% accuracy), emphasizing the efficacy of our representation learning methodology. This paper introduces a novel self-supervised methodology based on transformer architecture for well log curve representation learning. The method automates information extraction without requiring logging expertise and substantially enhances downstream machine learning model performance.
How to cite: Li, P., Zhou, J., Liu, Y., Zhang, J., Li, G., and Zhou, Y.: An Adaptive Masking Time Series Transformer based Representation Learning Model for Well Log Curves, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-1454, https://doi.org/10.5194/egusphere-egu25-1454, 2025.
Dimensionality reduction techniques have been successfully applied in remote sensing to reduce redundant information. However, achieving dimensionality reduction and lossless recovery for multispectral data at any global location remains a challenge, particularly given the complex and variable nature of surface conditions. Furthermore, it is still unclear if the reduced features maintain temporal continuity and can be effectively integrated with existing time series algorithms for disturbance detection. This study trains a Uniform Manifold Approximation and Projection (UMAP) model based on Harmonized Landsat Sentinel-2 (HLS) imagery to accomplish multispectral dimensionality reduction. Subsequently, the manifold embeddings are used in the Continuous Change Detection and Classification (CCDC) algorithm for land disturbance detection. Two key conclusions are drawn from this study: 1) a general multispectral dimensionality reduction model was constructed based on UMAP, which is applicable to all global land surfaces and any seasons. The manifold embeddings exhibit a stable value range and preserve the coherence of the time series. 2) compared to full-spectrum multispectral data, the manifold embeddings achieved comparable performance in image prediction and disturbance detection. Our study demonstrates the potential of manifold learning-based representation of global land surface reflectance spectra for lightweight storage and processing of dense satellite image time series, while keeping the ability to detect any kinds of land disturbance.
How to cite: Li, M. and Qi, J.: Manifold Embeddings for Multispectral Time-Series Land Disturbance Detection, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5726, https://doi.org/10.5194/egusphere-egu25-5726, 2025.
Causal Inference is essential for identifying and quantifying causal relationships in systems where randomized controlled experiments are infeasible, but the high dimensionality and co-variability structure of spatiotemporal dynamical systems such as the climate system pose special challenges for causal effect estimation. It is standard for climate scientists to reduce the dimension of their data with pre-processing procedures such as regional means and principal component analysis, but taking a regional mean may mask differences in spatial pattern – such as the difference between Eastern Pacific and Central Pacific El Niño events – that may be relevant for causal relationships. Similarly, principal component analysis may obscure true causal relationships because the spatial pattern associated with maximum co-variability may not be the causally relevant information. Instead of using these preprocessing techniques, the basic procedure of time series causal effect estimation can be simply extended to multivariate time series, but this introduces new complications and heightens already existing complications of time series causal effect estimation. Here, we discuss these complications and present practical solutions. Complications for multi-variate as well as univariate time series include: (1) neighboring points in time and space may be very similar if the scale of the spatiotemporal sampling rate is small relative to the characteristic scale of the variance, resulting in unstable estimations, (2) the do-calculus expression for estimating the response to a hard intervention may include calculations with spatiotemporal gradients so strong they would result in instabilities in the system, and finally, (3) it is often not possible to actually perform a hard intervention in dynamical systems, making the interpretation of the causal effect unclear. The first complication may be addressed using L2 regularization, and the second and third complications may be addressed by focusing on soft interventions of reasonable magnitude that approach zero on their spatiotemporal boundaries. A unique complication of multi-variate causal effect estimation is that, when using L2 regularization, the total causal influence of a climate variable will be penalized inverse-proportionally to the number of spatial datapoints. This complication can be addressed by scaling variables so that the total spatiotemporal variance, rather than the component-wise variance, is one. We showcase the power of the technique by quantifying the spatiotemporal causal effect of El Niño-related sea surface temperature variability on atmospheric pressure variability in the North Atlantic in unforced Community Earth System Model simulations. We demonstrate that spatiotemporal causal effect estimation allows us to simultaneously determine the relevant spatial patterns and more accurately quantify a pattern-dependent causal effect between ENSO and NAO that has thus far proven difficult to measure in observational studies.
How to cite: Herman, R. and Runge, J.: Spatiotemporal Causal Effect Estimation in Complex Dynamical Systems, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-6863, https://doi.org/10.5194/egusphere-egu25-6863, 2025.
Natural magnetic field measurement is essential for discovering fundamental physical mechanisms in space. Both the CSES mission and the DEMETER satellite equipped with the search coil magnetometer to observe the magnetic field waves. The CSES mission’s search coil magnetometer was developed by the School of Space Sciences Department of Beihang University. But the accuracy of these measurements is often degraded by artificial interference from reaction wheels on satellites. These wheels produce complex harmonic interference, often overlapping with the natural signal in both time and frequency domain, which makes it difficult to observe natural signals.
Traditional methods usually construct filters to separate interference. Advanced signal technologies have focused on reducing interference using self-adaptive signal decomposition methods in either time or frequency domain. In this field, Finley and Robert have used singular spectrum analysis to remove interference from in situ magnetic field data from the CASSIOPE/Swarm-Echo mission. But they did not settle the time-frequency overlap problem. In fact, most signal decomposition methods do not work well. These methods usually damage the natural signal because the overlapping areas remain indistinguishable.
In this paper, a novel method named the Instantaneous Phase Discontinuity (IPD) method is proposed to address this issue. Based on the sensitivity of instantaneous phase to variation of signal frequency, this method utilizes the discontinuities in the phase function to identify overlapping time-frequency regions. Subsequently, the natural signal within the overlapping region is carefully separated through frequency band contraction and envelope correction. IPD holds broad application prospects. As an example, IPD effectively separates interference from the time-frequency overlapping regions while preserving the integrity of natural signals when applied to data obtained from the CSES mission.
How to cite: Shi, F., Zeng, L., and Fu, Y.: An Innovative Technique for Reaction wheel Interference Separation in Satellite Magnetic Field Signals, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11558, https://doi.org/10.5194/egusphere-egu25-11558, 2025.
Posters virtual: Fri, 2 May, 14:00–15:45 | vPoster spot 4
EGU25-17965 | ECS | Posters virtual | VPS20
Lagrangian Evolution of the Trapping Capacity of Mesoscale Eddies in the Canary Eddy Corridor: A Numerical Modeling ApproachFri, 02 May, 14:00–15:45 (CEST) vPoster spot 4 | vP4.17
The Canary Eddy Corridor is a dynamic region of mesoscale eddy activity, playing a critical role in the transport of physical properties (heat and salt) and biogeochemical properties (nutrients, larvae, plankton) in the eastern North Atlantic. This study investigates the Lagrangian evolution of the trapping capacity of mesoscale eddies according to their lifecycle phases and vertical structure (surface vs. subsurface eddies).
We combine OceanParcels (an open-source Python toolbox) and an eddy identification and tracking algorithm with the GLORYS12V1 reanalysis product and altimetry data from AVISO to simulate particle release and track trajectories within eddies. Applying the eddy tracking algorithm at surface and subsurface levels in GLORYS12V1 reveals that subsurface eddies with a surface signal exhibit subsurface rotational velocities at the eddy core that occasionally exceed those of surface eddy cores. This highlights the potential misrepresentation of eddy transport capacity when relying solely on altimetry data, without accounting for the vertical structure, which can be better resolved through a combination of model outputs and observational data, such as non-standard Argo float configurations. Furthermore, a detailed analysis of the eddy lifecycle phases shows that mature eddies exhibit substantially greater trapping depths compared to their growth and decay stages. These findings align with earlier modeling analyses of dipoles originating south of Madagascar, which also highlight enhanced trapping depths in mature eddies.
The results provide a comprehensive view of the trapping capacity of mesoscale eddies throughout their lifecycle and vertical structure, emphasizing their critical role in biophysical coupling, ecological connectivity, and the transport of biogeochemical properties, as well as microplastics and other pollutants.
Acknowledgments: The first author is grateful for the internship grants ERASMUS +, AMI-MESRI, and TIGER.
How to cite: Vacca, D., Aguiar-González, B., and Morris, T.: Lagrangian Evolution of the Trapping Capacity of Mesoscale Eddies in the Canary Eddy Corridor: A Numerical Modeling Approach, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17965, https://doi.org/10.5194/egusphere-egu25-17965, 2025.
