Explore the vast range of titles available.

The scattering of charged particles by oxygen ion cyclotron harmonic (OCH) waves in the inner magnetosphere is investigated by evaluating the relevant quasi‐linear diffusion coefficients. Recent studies demonstrated that OCH waves are oxygen ion Bernstein modes and their complex kinetic dispersion relation has made it challenging to assess their role in scattering charged particles. The present study calculates the quasi‐linear diffusion coefficients for the scattering of electrons and ions by OCH waves using their kinetic dispersion relation. The results show that OCH waves can effectively scatter electrons between ∼100 eV and 100s keV via Landau resonance. They are also capable of heating cold helium and oxygen ions through cyclotron resonances. Specially, it is found that the 4th harmonic of OCH waves can lead to effective heating of helium ions, while oxygen ions would interact more efficiently with lower harmonics of OCH waves. Plain Language Summary Oxygen ion cyclotron harmonic (OCH) waves observed in the inner magnetosphere often have multiple spectral peaks at harmonics of the local oxygen ion cyclotron frequency. They have been shown to be excited by hot oxygen ion loss‐cone or ring/ring‐like distributions and follow a complicated kinetic dispersion relation for oxygen ion Bernstein waves. Since OCH waves cannot be described by the relatively simple cold plasma dispersion relation, it has been difficult to calculate their diffusion coefficients in scattering charged particles in quasi‐linear theory. The present study numerically solves the kinetic dispersion relation for OCH waves and then uses the results to calculate the corresponding quasi‐linear diffusion coefficients for electrons and ions. The diffusion coefficients obtained show that OCH waves can effectively interact with ∼100 eV to 100s keV electrons and are capable of heating cold helium and oxygen ions. Thus, OCH waves have their own unique contribution to the particle dynamics in the inner magnetosphere. Key Points Quasi‐linear diffusion coefficients are evaluated for particle scattering by oxygen ion cyclotron harmonic (OCH) waves for the first time OCH waves can scatter electrons in a wide energy range (∼100 eV–100s keV) via Landau resonance OCH waves are capable of heating cold helium and oxygen ions through cyclotron resonance

We discuss the situations under which Brownian yet non-Gaussian (BnG) diffusion can be observed in the model of a particle's motion in a random landscape of diffusion coefficients slowly varying in space (quenched disorder). Our conclusion is that such behavior is extremely unlikely in the situations when the particles, introduced into the system at random at t = 0, are observed from the preparation of the system on. However, it indeed may arise in the case when the diffusion (as described in Ito interpretation) is observed under equilibrated conditions. This paradigmatic situation can be translated into the model of the diffusion coefficient fluctuating in time along a trajectory, i.e. into a kind of the 'diffusing diffusivity' model.

Chorus waves are intense electromagnetic emissions critical in modulating electron dynamics. In this study, we perform two‐dimensional particle‐in‐cell simulations to investigate self‐consistent wave‐particle interactions with oblique chorus waves. We first analyze the electron dynamics sampled from cyclotron and Landau resonances with waves, and then quantify the advection and diffusion coefficients through statistical studies. It is found that phase‐trapped cyclotron resonant electrons satisfy the second‐order resonance condition and gain energy from waves. While phase‐bunched cyclotron resonant electrons cannot remain in resonance for long periods. They transfer energy to waves and are scattered to smaller pitch angles. Landau resonant electrons are primarily energized by waves. For both types of resonances, advection coefficients are greater than diffusion coefficients when the wave amplitude is large. Our study highlights the important role of advection in electron dynamics modulation resulting from nonlinear wave‐particle interactions. Plain Language Summary Wave‐particle interactions can modulate electron distributions through advection and diffusion. Previous studies focusing on advection and diffusion primarily relied on test particle simulations, which uses a simplified model of wave evolution. In this study, we perform self‐consistent simulations to investigate the wave‐particle interactions with chorus waves and quantify the advection and diffusion coefficients of resonant electrons. It is found that advection coefficients are greater than diffusion coefficients in both cyclotron and Landau resonances, indicating the significant role of nonlinear wave‐particle interactions. The quantification of advection and diffusion coefficients in a self‐consistent system is important for understanding and predicting the loss and energization processes in radiation belt electrons. This study complements previous diffusion models that regarded the evolution of electron dynamics in wave‐particle interactions as a slow diffusive process. Key Points Electron advection and diffusion in wave‐particle interactions with chorus waves are investigated through self‐consistent simulations The second‐order time derivative of gyrophase angle is nearly zero for phase‐trapped electrons but is negative for phase‐bunched electrons The advection and diffusion coefficients for cyclotron and Landau resonant electrons interacting with chorus waves are quantified

Oceanographic observations are limited by sampling rates, while ocean models are limited by finite resolution and high viscosity and diffusion coefficients. Therefore, both data from observations and ocean models lack information at small and fast scales. Methods are needed to either extract information, extrapolate, or upscale existing oceanographic data sets, to account for or represent unresolved physical processes. Here we use machine learning to leverage observations and model data by predicting unresolved turbulent processes and subsurface flow fields. As a proof of concept, we train convolutional neural networks on degraded data from a high‐resolution quasi‐geostrophic ocean model. We demonstrate that convolutional neural networks successfully replicate the spatiotemporal variability of the subgrid eddy momentum forcing, are capable of generalizing to a range of dynamical behaviors, and can be forced to respect global momentum conservation. The training data of our convolutional neural networks can be subsampled to 10–20% of the original size without a significant decrease in accuracy. We also show that the subsurface flow field can be predicted using only information at the surface (e.g., using only satellite altimetry data). Our results indicate that data‐driven approaches can be exploited to predict both subgrid and large‐scale processes, while respecting physical principles, even when data are limited to a particular region or external forcing. Our in‐depth study presents evidence for the successful design of ocean eddy parameterizations for implementation in coarse‐resolution climate models. Plain Language Summary Models of the ocean and ocean observations are imperfect. Due to this imperfection, simulations of the ocean and our observations are not quite the same as the true ocean currents. We, therefore, need ways to make our ocean data more realistic and complete and to make it more similar to the actual ocean. Scientists have traditionally approached this problem in a pen‐and‐paper style, considering physical theories and mechanisms. This study instead uses machine learning, which focuses on data as opposed to equations on a black board. We successfully use a particular type of machine learning algorithm, called a convolutional neural network, to make the most of current oceanographic data. This type of neural network works well even if ocean data are limited to a particular area. Future work will involve combining machine learning with physical theories of the ocean. Key Points We successfully use convolutional neural networks to predict unresolved turbulent processes and subsurface velocities The neural networks generalize to different regions, dynamical regimes, and forcing Global momentum conservation for eddy parameterization can be respected without sacrificing accuracy

Using the statistical wave power spectral profiles obtained from CRRES wave data within the 0000–0600 MLT sector under different levels of geomagnetic activity and a modeled latitudinal variation of wave normal angle distribution, we examine quantitatively the effects of lower band and upper band chorus on resonant diffusion of plasma sheet electrons for diffuse auroral precipitation in the inner magnetosphere. Whistler mode chorus‐induced resonant scattering of plasma sheet electrons is geomagnetic activity dependent, varying from above the strong diffusion limit (timescale of an hour) during active times (AE* > 300 nT) with peak wave amplitudes of >50 pT to weak scattering (timescale of a day) during quiet conditions (AE* < 100 nT) with typical wave amplitudes of ≤10 pT. Chorus waves present at different magnetic latitudes make distinct contributions to the net diffusion rates of plasma sheet electrons, largely depending on the latitudinal variation of wave power. Upper band chorus is the controlling scattering process for electrons from ∼100 eV to ∼2 keV, and lower band chorus is most effective for precipitating the higher energy (>∼2 keV) plasma sheet electrons in the inner magnetosphere. Efficient scattering by the combination of active time lower band and upper band chorus can cover a wide energy range from ∼100 eV to >100 keV and a broad interval of equatorial pitch angle, thereby accounting for the formation of observed electron pancake distribution. Decreased chorus scattering during less disturbed times can also modify the magnetic local time distribution of plasma sheet electrons. Compared to the effects of chorus waves, electron cyclotron harmonic wave‐induced resonant diffusion coefficients are at least 1 order of magnitude smaller and are negligible under any geomagnetic condition, indicating that chorus waves act as the major contributor dominantly responsible for diffuse auroral precipitation in the inner magnetosphere. Chorus‐driven momentum diffusion and mixed diffusion are also important. Lower band and upper band chorus can cause strong momentum diffusion of plasma sheet electrons in the energy ranges of ∼500 eV to ∼2 keV and ∼2 keV to ∼3 keV, respectively, which can significantly result in energization of the electrons and attenuation of the waves. Key Points Chorus can cause both efficient pitch angle scattering and momentum diffusion Chorus dominates over ECH waves to account for diffuse auroral precipitation Chorus scattering can also explain the formation of electron pancake distribution

ObjectiveTo develop and evaluate the performance of U-Net for fully automated localization and segmentation of cervical tumors in magnetic resonance (MR) images and the robustness of extracting apparent diffusion coefficient (ADC) radiomics features.MethodsThis retrospective study involved analysis of MR images from 169 patients with cervical cancer stage IB–IVA captured; among them, diffusion-weighted (DW) images from 144 patients were used for training, and another 25 patients were recruited for testing. A U-Net convolutional network was developed to perform automated tumor segmentation. The manually delineated tumor region was used as the ground truth for comparison. Segmentation performance was assessed for various combinations of input sources for training. ADC radiomics were extracted and assessed using Pearson correlation. The reproducibility of the training was also assessed.ResultsCombining b0, b1000, and ADC images as a triple-channel input exhibited the highest learning efficacy in the training phase and had the highest accuracy in the testing dataset, with a dice coefficient of 0.82, sensitivity 0.89, and a positive predicted value 0.92. The first-order ADC radiomics parameters were significantly correlated between the manually contoured and fully automated segmentation methods (p < 0.05). Reproducibility between the first and second training iterations was high for the first-order radiomics parameters (intraclass correlation coefficient = 0.70–0.99).ConclusionU-Net-based deep learning can perform accurate localization and segmentation of cervical cancer in DW MR images. First-order radiomics features extracted from whole tumor volume demonstrate the potential robustness for longitudinal monitoring of tumor responses in broad clinical settings.SummaryU-Net-based deep learning can perform accurate localization and segmentation of cervical cancer in DW MR images.Key Points• U-Net-based deep learning can perform accurate fully automated localization and segmentation of cervical cancer in diffusion-weighted MR images.• Combining b0, b1000, and apparent diffusion coefficient (ADC) images exhibited the highest accuracy in fully automated localization.• First-order radiomics feature extraction from whole tumor volume was robust and could thus potentially be used for longitudinal monitoring of treatment responses.

Lattice diffusion coefficients and partition coefficients have been determined for Li, Mg, Al, Sc, Ti, Cr, V, Mn, Co, Ni, Cu, Zn, Ga, Y, Zr, Nb, Mo, In, Lu, Hf, Ta and U in single crystals of natural magnetite as a function of oxygen fugacity (fO2) at 1150 °C and 1 bar by equilibration with a synthetic silicate melt reservoir. Most experiments were run for twelve hours, which was sufficient to generate diffusion profiles from 25 to > 1000 µm in length. The results were checked at one condition with two additional experiments at 66.9 and 161 h. The profiles were analysed using scanning laser-ablation inductively-coupled-plasma mass-spectrometry. Diffusion coefficients (D) were calculated by fitting data from individual element diffusion profiles to the conventional diffusion equation for one-dimensional diffusion into a semi−infinite slab with constant composition maintained in the melt at the interface. Equilibrium magnetite/melt partition coefficients are given by the ratio of the interface concentrations to those in the melt. Plots of log D as a function of log fO2 produce V-shaped trends for all the investigated elements, representing two different mechanisms of diffusion that depend on (fO2)−2/3 and (fO2)2/3. Diffusion coefficients at a given fO2 generally increase in the order: Cr < Mo ≈ Ta < V < Ti < Al < Hf ≈ Nb < Sc ≈ Zr ≈ Ga < In < Lu ≈ Y < Ni < U ≈ Zn < Mn ≈ Mg < Co < Li < Cu. Thus, Cu contents of magnetites are most susceptible to diffusive reequilibration, whereas the original content of Cr should be best preserved.

Vertical mixing parameterizations in ocean models are formulated on the basis of the physical principles that govern turbulent mixing. However, many parameterizations include ad hoc components that are not well constrained by theory or data. One such component is the eddy diffusivity model, where vertical turbulent fluxes of a quantity are parameterized from a variable eddy diffusion coefficient and the mean vertical gradient of the quantity. In this work, we improve a parameterization of vertical mixing in the ocean surface boundary layer by enhancing its eddy diffusivity model using data‐driven methods, specifically neural networks. The neural networks are designed to take extrinsic and intrinsic forcing parameters as input to predict the eddy diffusivity profile and are trained using output data from a second moment closure turbulent mixing scheme. The modified vertical mixing scheme predicts the eddy diffusivity profile through online inference of neural networks and maintains the conservation principles of the standard ocean model equations, which is particularly important for its targeted use in climate simulations. We describe the development and stable implementation of neural networks in an ocean general circulation model and demonstrate that the enhanced scheme outperforms its predecessor by reducing biases in the mixed‐layer depth and upper ocean stratification. Our results demonstrate the potential for data‐driven physics‐aware parameterizations to improve global climate models. Plain Language Summary The upper region of the ocean is highly energetic and is responsible for transferring mass, energy and biogeochemical tracers between the atmosphere and the deeper regions of the ocean. This transport takes place because of turbulent swirling motions, which are found to be of varying sizes. Climate models cannot represent all of these motions because smaller‐scale swirls are complex and require additional computational resources. As we cannot neglect those small swirls, we try to approximate their effects on larger‐scale motions using mathematical models. These models have a few ad hoc or empirical assumptions that lead to uncertainty when these climate models are used to project the future climate. To reduce this uncertainty, we augment an existing model of turbulent swirling process with machine learning, which replaces some ad hoc approximations with data‐driven neural networks. Neural networks can learn those missing processes more accurately than a traditional physics‐based model. The neural networks are shown to improve physics in climate simulations. Although we only touch on one component in an ocean climate model, this approach can be replicated to improve any other component that was using ad hoc assumptions and replace them with data‐driven models using techniques from machine learning. Key Points We improve a parameterization of vertical mixing in the ocean surface boundary layer using neural networks Neural networks are trained to predict the diffusivity of second moment closure and maintain energetic constraints of the original parameterization The improved scheme reduces biases of mixed layer depth and thermocline in an atmospherically forced ocean model

Chorus waves induce both electron acceleration and loss. In this letter, we provide significantly improved models of electron lifetime due to interactions with chorus waves. The new models fill the gap that previous models have on some magnetic local time (MLT) sectors of the Earth's magnetosphere. This improvement is critical for modeling studies. The lifetime models developed using two different methods are valid for electrons with an energy range from 1 keV to 2 MeV. To facilitate the integration of these new models into different ring current and radiation belt codes, we parameterize the electron lifetime as a function of L$L$ ‐shell and electron kinetic energy at each MLT and geomagnetic activity (Kp). The parameterized electron lifetimes exhibit strong dependencies on L$L$ ‐shell, MLT, and energy. Simulations using these new models demonstrate improved agreement with satellite observations compared to simulations using previous models, advancing our understanding of electron dynamics in the magnetosphere. Plain Language Summary There are a large number of energetic electrons trapped by our Earth's magnetic field in the near‐Earth space. The regions populated by these high energy electrons are called ring current and radiation belts. It is important to understand the dynamics of the energetic electrons because they can be dangerous to satellites and astronauts flying through these regions. Electromagnetic waves in these regions play an important role in the dynamic of ring current and radiation belt electrons. Among these waves, whistler mode chorus wave is an important wave that can cause both acceleration and loss of the energetic electrons. In our previous studies, we calculated diffusion coefficients to quantify the effect of chorus waves on the energetic electrons. Based on these diffusion coefficients, in this study, we estimate the lifetime of energetic electrons due to their interactions with chorus waves. To make this lifetime model more convenient to be used in different ring current and radiation belt models, we apply polynomial fits to the calculated lifetime. Our new lifetime model is more advanced than previous models, especially in the space coverage. We test the new models in simulations and the results agree better with satellite observations than the previous models do. Key Points The new lifetime model provides extended space coverage in comparison to current widely used lifetime models Such parameterized lifetimes are very significant for simulations of the dynamics of radiation belt and ring current electrons Using the new electron lifetime model in simulations improves the agreement between the simulation results and the satellite observations

Energetic particle (EP) transport induced by the microturbulence are studied both theoretically and numerically. Based on the quasi-linear theory, the expressions of the four diffusion coefficients describing the EP transport in radial-velocity space (r, v) are derived: the radial diffusion coefficient Drr, velocity diffusion coefficient Dvv, and two cross-terms Drv and Dvr. It is found that Drr scales as E−2 for the deeply trapped particles (TPs), as E−3/2 for the normally passing particles (PPs), and as E−1 for the purely PPs, where E is the EP energy. The normalized D^rv and D^vv scale, respectively, as E−3/2 and E−2 for the purely PPs, and both as E−2 for the deeply TPs. To verify the analytical results, the EP transport induced by the ion temperature gradient turbulence is simulated using GTC. The energy scaling of the simulated diffusion coefficients agrees well with the theoretical predictions. To reveal the underlying physical mechanisms, the finite Larmor radius (FLR) effects, finite orbit width (FOW) effects, and wave-particle resonance condition contributions are investigated separately. The results show that the FLR effects contribute an E−1/2 scaling for the TPs and normally PPs, FOW effects contribute an E−1/2 scaling for both the deeply TPs and PPs, and resonance condition contributes, respectively, an E−1 scaling for the TPs and an E−1/2 scaling for the PPs, consistent with the theoretical analyses. Apart from the energy dependence, the dependence of Drr on the EP pitch is also studied. With the increase of the pitch, the energy scaling is found to exhibit a continuous variation, providing a possible explanation for the discrepancy in previous works on the EP transport energy scaling. Finally, a comparison of the diffusion coefficients magnitudes is made, indicating that Drr dominates the EP transport.