Search for: All records

Abstract Broadband very low frequency (VLF, 3–30 kHz) energy emitted from lightning travels long distances in subionospheric propagation modes. Some of this energy couples into the ionosphere and travels to satellite altitudes along a predominantly field‐aligned path in what is known as a “whistler” propagation mode. Along this path, these VLF signals (known as whistlers) encounter both frequency‐dependent absorption and dispersion. When observed by satellites, these signals can be used to estimate the electron density profile along the path from the whistler to the satellite. A new technique for estimating electron density profiles in both the lower and upper regions of the ionosphere is presented using trans‐ionospheric VLF whistlers recorded by the DEMETER satellite. This technique employs an inverse modeling‐based approach to select a parameterized model of electron densities in the D‐region (60–90 km), as well as a scaling term that can be applied to the electron density profiles provided by the International Reference Ionosphere (IRI) for altitudes above the D‐region (90–700 km). Leveraging a set of 52 whistlers recorded by the DEMETER satellite in 2010, the algorithm produced consistent estimates for parameters representing the altitude and sharpness of the D‐region, as well as the altitude and thickness of an often‐occurring shelf in the D‐region, all of which are in alignment with known values for these parameters. Additionally, the algorithm produced a consistent scaling term for the IRI model of the E and F regions for the 52 recorded whistlers, with a mean value of 0.507 and a variance of only 0.009. 

Abstract Sporadic‐E is an ionospheric phenomenon defined by strong layers of plasma which may interfere with radio wave propagation. In this work, we develop deep learning models to improve the understanding of , including the presence, intensity and height of the layers. We developed three separate models. The first, building off earlier work in (J. A. Ellis et al., 2024,https://doi.org/10.1029/2023sw003669), includes only the main features from radio occultation (RO) measurements. The second adds to that time, date, location, geomagnetic and solar indices, solar winds, x‐ray flux, weather and lightning. A third model excludes RO measurements but includes the rest. In training the first two models, the ordinary mode critical frequency (foEs), a measure of intensity, and height (hEs) parameters extracted from ionosondes were used as the “ground truth” target variables. In training the third model, estimated foEs and hEs values from the RO model were added as target variables to augment the data set and produce physically reasonable model predictions globally. We find that the second model performs well with intensity prediction tasks, but struggles with height estimations, which is likely due to the tangent point assumption made during RO signal processing and errors inherent in the ionosonde extracted virtual heights. The third model performed reasonably well considering the lack of in situ RO measurement. The third model performs the best on height predictions, which points to the height being very climatologically driven, whereas the intensity is a more complex interaction of several variables. 

Abstract We model the electron density in the topside of the ionosphere with an improved machine learning (ML) model and compare it to existing empirical models, specifically the International Reference Ionosphere (IRI) and the Empirical‐Canadian High Arctic Ionospheric Model (E‐CHAIM). In prior work, an artificial neural network (NN) was developed and trained on two solar cycles worth of Defense Meteorological Satellite Program data (113 satellite‐years), along with global drivers and indices to predict topside electron density. In this paper, we highlight improvements made to this NN, and present a detailed comparison of the new model to E‐CHAIM and IRI as a function of location, geomagnetic condition, time of year, and solar local time. We discuss precision and accuracy metrics to better understand model strengths and weaknesses. The updated neural network shows improved mid‐latitude performance with absolute errors lower than the IRI by 2.5 × 10⁹to 2.5 × 10¹⁰e⁻/m³, modestly improved performance in disturbed geomagnetic conditions with absolute errors reduced by about 2.5 × 10⁹ e⁻/m³at high Kp compared to the IRI, and high Kp percentage errors reduced by >50% when compared to E‐CHAIM. 

Abstract In this work, convolutional neural networks (CNN) are developed to detect and characterize sporadic E (E_s), demonstrating an improvement over current methods. This includes a binary classification model to determine ifE_sis present, followed by a regression model to estimate theE_sordinary mode critical frequency (foEs), a proxy for the intensity, along with the height at which theE_slayer occurs (hEs). Signal‐to‐noise ratio (SNR) and excess phase profiles from six Global Navigation Satellite System (GNSS) radio occultation (RO) missions during the years 2008–2022 are used as the inputs of the model. Intensity (foEs) and the height (hEs) values are obtained from the global network of ground‐based Digisonde ionosondes and are used as the “ground truth,” or target variables, during training. After corresponding the two data sets, a total of 36,521 samples are available for training and testing the models. The foEs CNN binary classification model achieved an accuracy of 74% and F1‐score of 0.70. Mean absolute errors (MAE) of 0.63 MHz and 5.81 km along with root‐mean squared errors (RMSE) of 0.95 MHz and 7.89 km were attained for estimating foEs and hEs, respectively, when it was known thatE_swas present. When combining the classification and regression models together for use in practical applications where it is unknown ifE_sis present, an foEs MAE and RMSE of 0.97 and 1.65 MHz, respectively, were realized. We implemented three other techniques for sporadic E characterization, and found that the CNN model appears to perform better.