Local vortex-structured auroral spiral and a large-scale transpolar arc (TPA) were contemporaneously observed by the Polar ultraviolet imager (UVI), when a substorm almost recovered. The TPA grew along the dawnside auroral oval from the nightside to the dayside (oval-aligned TPA), and a chain of multiple auroral spots and spiral were located azimuthally near the poleward edge of the nightside auroral oval. Contemporaneous appearances of the TPA and the auroral spiral can be seen after the spiral appeared alone. Polar also detected the oval-aligned TPA and another dawnside TPA with the nightside end distorted toward the premidnight sector (J-shaped TPA) before and after the spiral’s formation, respectively. To examine these associated magnetotail structures, we performed global magnetohydrodynamic (MHD) simulations, based on two different types of code, BAT-S-RUS and improved REPPU, and examined how the field-aligned current (FAC) profiles varied in association with changes of the auroral form to TPA and/or auroral spiral. Global MHD simulations with the two different types of code can reproduce the TPAs and the associated FAC structures in the magnetotail. The auroral spiral and its nightside FAC profile, however, were not formed in both simulations, suggesting that its formation process cannot be treated within an MHD framework but is closely related to some kinetic process. When the J-shaped TPA and the auroral spiral contemporaneously appeared, the two MHD simulations could not reproduce the TPA, spiral and their associated magnetotail FAC structures, also advocating that a kinetic effect related to the spiral formation might prevent the TPA occurrence.

Results from a study of automatic aurora classification using machine learning techniques are presented. The aurora is the manifestation of physical phenomena in the ionosphere-magnetosphere environment. Automatic classification of of auroral images from the Arctic and Antarctic is therefore an attractive tool for developing auroral statistics and for supporting scientists to study auroral images in an objective, organized and repeatable manner. Although previous studies have presented tools for detecting aurora, there has been a lack of tools for classifying aurora into subclasses with a high precision (>90%). This work considers seven auroral subclasses; breakup, colored, arcs-bands, discrete, patchy, edge and clear-faint. Five different deep neural network architectures have been tested along with the well known classification algorithms; k nearest neighbor (KNN) and a support vector machine (SVM). A set of clean nighttime color auroral images, without ambiguous auroral forms, moonlight, twilight, clouds etc., were used for training and testing. The deep neural networks generally outperformed the KNN and SVM methods, and the ResNet-50 architecture achieved the highest performance with an average classification precision of 92%. Although the results indicate that high precision aurora classification is an attainable objective using deep neural networks, it is stressed that a common consensus of the auroral morphology and the criteria for each class needs to be established before classification of ambiguous images can be readily achieved.