The prediction of post-sunset equatorial plasma depletions (EPDs), often called ionospheric plasma bubbles, has remained a challenge for decades. In this study, we introduce the Ionospheric Bubble Probability (IBP), an empirical model predicting the occurrence probability of EPDs derived from 9 years of CHAMP and 8.5 years of Swarm magnetic field measurements. The model predicts the occurrence probability of EPDs for a given longitude, day of year, local time and solar activity, for the altitude range 350-500 km, and low geographic latitudes of ± 45◦. IBP has been found to successfully reconstruct the distribution of EPDs as reported in previous studies from independent data. IBP has been further evaluated using one-year of partly untrained data of the Ionospheric Bubble Index (IBI). IBI is a Level 2 product of the Swarm satellite mission used for EPD identification. The relative operating characteristics (ROC) curve shows positive excursion above the no-skill line with Hanssen and Kuiper’s Discriminant (H&KSS) score of 0.66, 0.73, and 0.65 at threshold model outputs of 0.22, 0.18, and 0.18 for Swarm A, B, and C satellites, respectively. Additionally, the reliability plots show proximity to the diagonal line with a fairly decent Brier Skill Score (BSS) of 0.317, 0.320, and 0.316 for Swarm A, B, and C respectively. These tests indicate that the model performs significantly better than a no-skill forecast. The IBP model offers a compelling glimpse into the future of EPD forecasting, thus demonstrating its potential to reliably predict EPD occurrences. The IBP model is made publicly available.

Satellites with dual-frequency Global Navigation Satellite Systems (GNSS) receivers can measure integrated electron density, known as slant Total Electron Content (sTEC), between the receiver and transmitter. Precise relative variations of sTEC are achievable using phase measurements on L1 and L2 frequencies, yielding around 0.1 TECU or better. However, CubeSats like Spire LEMUR, with simpler setups and code noise in the order of several meters, face limitations in absolute accuracy. Their relative accuracy, determined by phase observations, remains in the range of 0.1-0.3 TECU. With a substantial number of observations and comprehensive coverage of lines of sight between Low Earth Orbit (LEO) and GNSS satellites, global electron density can be reconstructed from sTEC measurements. Utilizing 27 satellites from various missions, including Swarm, GRACE-FO, Jason-3, Sentinel 1/2/3, COSMIC-2, and Spire CubeSats, a cubic B-spline expansion in magnetic latitude, magnetic local time, and altitude is employed to model the logarithmic electron density. Hourly snapshots of the three-dimensional electron density are generated, adjusting the model parameters through non-linear least-squares based on sTEC observations. Results demonstrate that Spire significantly enhances estimates, showcasing exceptional agreement with in situ observations from Swarm and Defense Meteorological Satellite Program (DMSP). The model outperforms contemporary climatological models, such as International Reference Ionosphere (IRI)-2020 and the neural network-based NET model. Validation efforts include comparisons with ground-based slant TEC measurements, space-based vertical TEC from Jason-3 altimetry, and global TEC maps from the Center for Orbit Determination in Europe (CODE) and the German Research Center for Geosciences (GFZ).