Speaker: Randa Natras, Deutsches Geodätisches Forschungsinstitut der Technischen Universität München (DGFI-TUM)
Title: Investigation of PROBA2 LYRA data for predicting the Vertical Total Electron Content (VTEC) of the ionosphere with machine learning
Abstract:
The accuracy and reliability of Global Navigation Satellite Systems (GNSS) positioning and navigation can be affected by the state of the Earth’s ionosphere and space weather events. In order to minimize the refraction of the GNSS signals by the ionosphere, corresponding corrections from the spatially and temporally variable Vertical Total Electron Content (VTEC), including the solar and space weather impact on VTEC, need to be provided in a timely manner. Machine learning techniques are utilized to learn these nonlinear relationships from data in order to develop a model for VTEC forecasting. Within the PROBA2 Guest investigator program, the influence of solar irradiance on VTEC is studied using channel 4 (zirconium) and level 3 data from the solar instrument LYRA. Furthermore, LYRA observations are used as input to the VTEC prediction model, which is based on the eXtreme Gradient Boosting (XGBoost) learning method. Finally, the contribution of LYRA data to the VTEC forecast is analyzed. This presentation presents preliminary results of the applicability of LYRA data for VTEC prediction with machine learning.
Location: Meridian Room
Online: https://us06web.zoom.us/j/83412518381?pwd=dm5xNG5YT3ZLbDRYY1hvK3NwRml5Zz09