Funded thesis topic

Announcement transmitted by Aurélie Marchaudon (IRAP)

Thesis subject: New indices of magnetic activity to constrain wave propagation models in the ionosphere, application to SATIS

Keywords: ionosphere, geomagnetism, space plasma physics, Space Weather, deep learning

Co-funding secured: Thalès / Defense Agency

Supervisors: Aurélie Marchaudon (IRAP, Toulouse) and Aude Chambodut (EOST, Strasbourg)

Applications to be sent before the 7th of April 2019 to: aurelie.marchaudon@irap.omp.eu and aude.chambodut@unistra.fr

We currently face a growing need for a better representation of the terrestrial space environment (Magnetosphere-Ionosphere-Thermosphere system) and its response to solar events, responsible for magnetic storms. This need is critical in the context of Space Weather, whose goal is to predict the response of the space environment especially to constrain the propagation of HF and UHF waves in the ionosphere. In recent years, we have helped to improve this description of the terrestrial environment by proposing new indices of magnetic activity, called the alpha indices. Their main characteristics is to have a good temporal resolution of about 15 minutes. These indices are intended to supplant the historical Kp indices developed in the 1940s and 1960s, now obsolete, but which are still used to constrain the empirical models of the ionosphere in propagation tools such as SATIS.

In this PhD thesis, we propose to create a new generation of indices derived from the alpha indices, with a better spatial resolution via a subdivision in local time.

•    The first part of the thesis will be devoted to the construction of these new sectoral indices and their characterization during intense solar events, responsible for magnetic storms and ionospheric disturbances. To do this, we will develop an automated tool based on a neural network to characterize the events from the point of view of the source (solar wind) and from the point of view of the magnetospheric effect (indices of magnetic activity) and the ionospheric effect (irregularities causing loss of HF and UHF signals).

•    The second part will then be dedicated to the automatic detection of geoeffective events from these indices. In this second phase, we will evolve the automated tool by relying on Deep Learning techniques to make the detection more efficient and even possibly the prediction of the indices, in particular we will be interested in the potential precursors, in terms of types of events and physical parameters.

References:

Chambodut, A., A. Marchaudon, C. Lathuillère, M. Menvielle, and E.

Foucault (2015), New hemispheric geomagnetic indices α with 15 min time

resolution, J. Geophys. Res. Space Physics, 120, 9943–9958,

https://doi.org/10.1002/2015JA021479.

Chambodut, A., A. Marchaudon, M. Menvielle, F. El-Lemdani Mazouz, and C.

Lathuillère (2013), The K-derived MLT sector geomagnetic indices,

Geophys. Res. Lett., 40, 4808-4812, https://doi.org/10.1002/grl.50947.