Quand

20 novembre 2026    
9h00 - 12h00

Où

LPC2E
3 Av. de la Recherche Scientifique, Orléans, 45071

Type d’évènement

The PhD defense: “Characterizing space plasmas through multipoint analysis of data collected by a constellation of satellites” by Mehul Chakraborty will take place on Friday, 20 November 2026 at 09h00, in the LPC2E conference room , CNRS, Orleans. For those who cannot attend, here is a link to follow the presentation.

The thesis develops J-filtering, a new multipoint technique to detect electric current structures in space plasmas, estimate their geometrical properties and visualize their spatial distributions in three dimensions. The method uses the Maxwell-Ampere law to model candidate current structures and compares their magnetic fields with spacecraft measurements using Capon’s adaptive filter-bank approach. Its mathematical implementation is adapted from the k-filtering technique, and the same general estimator framework is used for different current models. Applications cover a range of current geometries and scales, from large-scale currents to localized Gaussian cylindrical currents and Harris current sheets. For large-scale currents, analysis of two Cluster magnetopause crossings shows agreement with the well established curlometer technique. Synthetic statistical tests also show that large-scale J-filtering is more stable than curlometer for planar spacecraft configurations. For localized cylindrical currents and current sheets, tests using synthetic measurements show that, when the current scale is estimated reasonably and the magnetic field is sampled appropriately, the current density structure can be recovered. Comparing the stability of reconstructed directions and the alignment between modeled and measured magnetic fields over the time the current structure crosses the spacecraft configuration provides a diagnostic for identifying the most probable current model. The thesis therefore extends magnetic-field-based current estimation beyond the large-scale regime of the curlometer, while retaining an explicit dependence on the assumed current structure and its sampling. It provides a basis for detecting, distinguishing and visualizing localized currents with data from all existing (Cluster, MMS) and future (HelioSwarm, Plasma Observatory) multipoint missions.

(Transmis par Mehul Chakraborty)