J/AJ/170/153 CNEOS fireball light curves: BLADE classification (Silber+, 2025)
BLADE: An automated framework for classifying light curves from the Center for
Near-Earth Object Studies fireball database.
Silber E.A., Sawal V.
<Astron. J., 170, 153 (2025)>
=2025AJ....170..153S 2025AJ....170..153S
ADC_Keywords: Photometry; Minor planets; Comets; Meteorites; Earth; Optical;
Infrared; Models
Keywords: Asteroids ; Meteoroids ; Light curves ; Light curve classification
Abstract:
Fireballs (bolides) are high-energy luminous phenomena produced when
meteoroids and small asteroids enter Earth's atmosphere at hypersonic
speeds, often resulting in fragmentation or complete disintegration
accompanied by significant energy release. The resulting bolide light
curves capture temporal brightness variations as these objects
traverse increasingly dense atmospheric layers, providing essential
information on meteoroid entry dynamics, fragmentation behavior, and
atmospheric energy deposition processes. The Center for Near-Earth
Object Studies' (CNEOS) continuously expanding fireball database
offers a globally comprehensive archive of bolide events, including
light curves and associated metadata. Events associated with
infrasound detections allow direct correlations between acoustic
signatures and light curve features, therefore enabling detailed
analyses of fragmentation dynamics and energy deposition. Here, we
introduce Bolide Light-curve Analysis and Discrimination Explorer
(BLADE), a robust and high-fidelity framework specifically designed to
analyze bolide light curves for objects detected from space. BLADE
incorporates a processing pipeline integrating Savitzky-Golay
filtering, prominence-based peak detection, and gradient analysis,
enabling systematic identification and classification of fragmentation
events and their associated energy release characteristics.
Preliminary results demonstrate that BLADE reliably distinguishes
distinct bolide behaviors, providing an objective, scalable
methodology for characterization and analysis of large bolide light
curve data sets. This foundational work establishes a novel pathway
for advanced bolide research, with promising applications in planetary
defense and global atmospheric monitoring. Future research should
adopt an integrative approach combining CNEOS optical data with
complementary infrasound measurements, further clarifying
relationships between bolide energy deposition and acoustic
signatures, thus refining our understanding of meteoroid and asteroid
atmospheric entry processes.
Description:
At the time of this analysis, the Center for Near-Earth Object
Studies' (CNEOS) fireball database, maintained by NASA's Jet
Propulsion Laboratory (JPL), contained approximately 850 light curves
documenting bolide events detected by U.S. Government (USG) sensors.
The initial subset included 144 events, each having infrasound
detections published in the literature. After careful evaluation of
data quality and completeness in both optical and acoustic records, the
final subset totaled 124 bolides. The light curve data were obtained
directly from the CNEOS database (http://cneos.jpl.nasa.gov/fireballs/).
Here, we use BLADE, an integrated framework comprising data
preprocessing, Savitzky-Golay filtering for noise reduction, peak
detection, gradient analysis, and event classification.
The CNEOS fireball database is widely regarded as one of the most
comprehensive observational archives available for studying meteoroid
entry events and their interaction with Earth's atmosphere. Managed
by NASA's JPL, the database aggregates observations from USG
sensors, which detect bolides on a global scale through their
radiative emissions in optical and infrared wavelengths.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table1.dat 139 124 List of events and output classifications
--------------------------------------------------------------------------------
See also:
B/comets : Database of the orbital elements of comets (Rocher, 2007)
J/AJ/156/60 : Thermal model fits for short-arc NEOs: NEOWISE (Masiero+, 2018)
J/A+A/667/A157 : 824 fireballs observed in 2017-2018 (Borovicka+, 2022)
J/A+A/679/A148 : Compositional properties of NEOs (Sergeyev+, 2023)
J/A+A/674/A50 : Near-Earth Object (NEO) properties (Deshapriya+, 2023)
J/A+A/682/A29 : Detection of the Yarkovsky effect on NEAs (Fenucci+, 2024)
J/A+A/704/A241 : Calibrated spectra of 31 fireballs (Vojacek+, 2025)
http://cneos.jpl.nasa.gov/fireballs/ : CNEOS fireball database (NASA JPL)
Byte-by-byte Description of file: table1.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 3 I3 --- Event [147/905] CNEOS event number
5- 14 A10 "Y/M/D" Date Bolide event date
16- 23 A8 --- Time Bolide event time in UTC
25- 29 F5.1 deg Lat [-83.7/88.5] Bolide latitude
31- 36 F6.1 deg Long [-179.3/174.2] Bolide longitude
38- 41 F4.1 km Alt [16.7/59.3]? Bolide altitude
43- 46 F4.1 km/s Vel [9.8/44.8]? Velocity
48- 52 F5.1 km/s xVel [-35.4/21.3]? Velocity in x-direction
54- 58 F5.1 km/s yVel [-43.5/23.4]? Velocity in y-direction
60- 64 F5.1 km/s zVel [-31.2/23.9]? Velocity in z-direction
66- 73 E8.2 J Radiated [2.8/3.75e14] Total radiated energy
75- 81 F7.3 --- Impact [0.073/440.0] Total impact energy (1)
83- 88 F6.2 deg Azimuth [6.89/358.98]? Azimuth
90- 94 F5.2 deg Angle [4.42/83.54]? Entry angle
96- 119 A24 --- Type Fragmentation type (2)
121- 124 I4 yr Year [2000/2022] UTC year
126- 127 I2 month Month UTC month
129- 130 I2 d Day UTC day
132- 133 I2 h Hr UTC hour
135- 136 I2 min Min UTC minute
138- 139 I2 s Sec UTC second
--------------------------------------------------------------------------------
Note (1): Expressed in units of kilotons (kt) of trinitrotoluene (TNT), where
1kt TNT = 4.184e12 J.
Note (2): Fragmentation type as follows:
Airburst = 9 occurrences
Continuous Fragmentation = 3 occurrences
Discrete Fragmentation = 81 occurrences
Single Peak = 31 occurrences
--------------------------------------------------------------------------------
History:
From electronic version of the journal
License: CC-BY-4.0
(End) Prepared by [AAS], Robin Leichtnam [CDS] 29-Jun-2026