Using the Geostationary Lightning Mapper data to calculate orbit and trajectory of bolide events and study fragmentation

Abstract

<p>An interesting new space-based data source for bolide detection has emerged within the past few years.  Called the Geostationary Lightning Mapper (GLM) instrument, it resides aboard both National Oceanic and Atmospheric Administration (NOAA)'s Geostationary Operational Environmental Satellite (GOES) 16 and 17, and has been shown to be a legitimate detector of bolide events across the contiguous United States (Jenniskens et al. 2018).  While it is clearly beneficial to be able to detect and verify bolide events using space-based remote sensing, the ability to use GLM data to build trajectory and orbital determinations for events would be an even greater advantage, as it would give us a uniquely independent view of the bolide event from space, and could add substantially to any pre-existing ground-based information available for the event.  We will analyze light curves constructed from GLM data to find about fragmentation of these bodies, which might provide novel insights into the nature of the materials that comprise bolides and the physics behind fragmentation.  We shall investigate these new avenues of approach-- orbit and trajectory calculations, and fragmentation analysis, both with GLM data. </p> <p>Trajectory and orbit for these events will be calculated using publicly available Level 2 data. Level 2 data is produced by the application of a reduction pipeline to the raw sensor data (Level-0) which reduces noise and glint from the image, and geolocates pixels which satisfy a threshold condition. These instruments provide calibrated pixel intensity and location data for events occurring above the continental United States at or above -14 magnitude at a 500-Hz sampling rate in 20-second intervals (Jenniskens et al. 2018). Individually triggered pixels are labelled 'events', and a mass-weighted centroid is applied to get the 'group' data, which has sub-pixel accuracy.  GLM data is released in NetCDF4 format with timestamped flash, group, and event position, in planetographic coordinates, and the calibrated luminous energy. With the stereo events, we can use line of sight to work out a trajectory.  Orbit is calculated by using the velocity vector measured at the uppermost part of the luminous flight of the object.  At the entry location, we calculate the heliocentric position and velocity vector and carry out backward time-integration using the open-source N-body software REBOUND (Rein and Liu 2012) to compute the position and velocity outside of Earth's Hill sphere.</p> <p>We will investigate if GLM data provides greater insight into not just trajectory and orbit calculations, but also fragmentation analysis.  It has been shown by Jenniskens et al. (2018) and Brown et al. (2019) that optical energy estimates can be reliably generated from GLM observations.  We will attempt to show how the GLM data can be further utilized to assist in fragmentation analysis by modeling the light curve.  Also, some of the events recorded by the GLM exhibit rapid fluctuations, or a "jitter" in the light curve just after peak intensity, which could be interpreted as a large-scale scattering of parent body material.  This is further evidenced by multiple smaller events being seen by the sensor at different locations during the same time stamp.  We have identified 42 such events and will continue to examine these unique light curves.  </p> <p>Ultimately, we investigate the extent to which GLM data can be used to get accurate trajectory and orbital calculations, in combination with ground based observations.  We will highlight the importance of being able to verify trajectories with the Spalding Allsky Camera Network and will explore possible benefits of accessing Level 0 (L0) unfiltered non-centroided data from the GLM instruments in order to be able to calculate a more robust trajectory.  We present preliminary results detailing the conditions under which a trajectory determination from GLM data is possible and fragmentation modeling of high-resolution events. </p>