Fig. 5.

Example visual comparison of the predictive performance of our physical model with that of the best competitor model for the dawn ionosphere. First row: ground truth ΔTEC overlaid on noisy draws from the ground truth which are the observations. Second and third rows: posterior mean with uncertainty contours for the physical model and best competitor model respectively. Fourth and fifth rows: residuals between posterior means and ground truth for the physical model and best competitor model respectively. First two columns: results for experimental conditions, (10 directions, 2.5 mTECU noise), for a central antenna (near to reference antenna) and a remote station (far from reference antenna). Last two columns: results for experimental conditions, (30 directions, 1.6 mTECU noise), for a central antenna and a remote station.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.