Moriba Jah, GaiaVerse Ltd., Jah Decision Intelligence Group, University of Texas at Austin; Anirudh Selvam, University of Texas at Austin
Keywords: NEXRAD Level II, radar data analysis, cloud classification, reentering bodies, space debris, Doppler velocity, spectral width, machine learning.
Abstract:
The ability to accurately distinguish between meteorological phenomena and artificial reentering bodies is crucial for atmospheric monitoring, aerospace safety, and space situational awareness. This paper presents a methodology for differentiating between clouds and reentering objects using Next Generation Weather Radar (NEXRAD) Level II data. Traditional weather radar systems are primarily designed for meteorological applications, detecting precipitation, cloud structures, and storm dynamics. However, these systems also register echoes from non-meteorological targets, including space debris and artificial objects undergoing atmospheric reentry. Misidentifying these objects as clouds can lead to false meteorological interpretations, whereas failing to detect them poses a risk to aerospace operations. Our approach leverages the rich dataset provided by NEXRAD Level II, which includes reflectivity, radial velocity, and spectrum width measurements at high temporal and spatial resolution. By analyzing key radar signatures, we establish a set of distinguishing characteristics between clouds and reentering bodies. Clouds typically exhibit consistent spatial coherence, lower Doppler velocities, and a gradual evolution of reflectivity patterns. In contrast, reentering objects produce transient, high-velocity signatures with irregular motion patterns, rapid linear re-entry trajectory and higher spectral width, indicative of turbulence and non-meteorological scatterers. To validate our methodology, we analyze case studies involving known space object reentries and compare their radar characteristics against typical cloud formations. The results demonstrate that our classification framework significantly improves detection accuracy, reducing false positives associated with clouds while correctly identifying reentering objects. These findings have implications for space debris tracking, air traffic safety, and meteorological data integrity. Future work will focus on integrating additional remote sensing datasets and enhancing real-time detection capabilities and incorporating a machine learning-based classification model trained on historical radar data, allowing for automated detection of anomalous signatures associated with reentry events.
Date of Conference: September 16-19, 2025
Track: Atmospherics/Space Weather