In this paper, regression-based deep convolutional neural networks (CNN), with 12 layers, are developed for predicting the maximal amplitude of the southward component of the near-Earth magnetic field from a passing interplanetary coronal mass ejection (ICME).

We have analyzed 33 cavities observed between 2012 and 2018, from solar activity maximum to minimum. For each cavity we applied a differential emission measure method to obtain both a temperature distribution and a value of the average temperature.

Gibson et al. have analyzed a long-lived cavity observed from 17 March 2012 to 21 March 2012. For this cavity we applied a differential emission measure method to obtain both a temperature distribution and a value of the average temperature during all five observational days.

The magnetic field in the corona is important for understanding solar activity. Linear polarization measurements inforbidden lines in the visible/IR provide information about coronal magnetic direction and topology.

Deriving the strength and direction of the three-dimensional (3D) magnetic field in the solar atmosphere is fundamental for understanding its dynamics. Volume information on the magnetic field mostly relies on photospheric and/or chromospheric surface vector magnetic fields.

Magnetic fields suspend the relatively cool material of solar prominences in an otherwise hot corona. A comprehensive understanding of solar prominences ultimately requires complex and dynamic models, constrained and validated by observations spanning the solar atmosphere.