Reprocessing the NEAT Dataset: Preliminary Results Journal: PSJ astropy ( Astropy Collaboration et al. 2022 ; Astropy Collaboration et al. 2018 ; Astropy Collaboration et al. 2013 ) , astroquery ( Ginsburg et al. 2019 ) , photutils ( Bradley et al. 2023 ) , sklearn ( Pedregosa et al. 2011 ) , Astrom
C. R. Nugent, J. M. Bauer, O. Benitez, M. Blain, N. D’Souza, S. Garimella, M. Goldwater, Y. Kim, H. C. G. Larsen, T. Linder, K. Mackowiak, Z. McGinnis, E. Pan, C. C. Pedersen, P. Sadhwani, F. Spoto, N. J. Tan, P. Vereš, C. Xue · 2025
Read Reprocessing the NEAT Dataset: Preliminary Results Journal: PSJ astropy ( Astropy Collaboration et al. 2022 ; Astropy Collaboration et al. 2018 ; Astropy Collaboration et al. 2013 ) , astroquery ( Ginsburg et al. 2019 ) , photutils ( Bradley et al. 2023 ) , sklearn ( Pedregosa et al. 2011 ) , Astrom by C. R. Nugent, J. M. Bauer, O. Benitez, M. Blain, N. D’Souza, S. Garimella, M. Goldwater, Y. Kim, H. C. G. Larsen, T. Linder, K. Mackowiak, Z. McGinnis, E. Pan, C. C. Pedersen, P. Sadhwani, F. Spoto, N. J. Tan, P. Vereš, C. Xue (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-20032.txt.
Introduction
Nugent and co-authors report in 2025 the first results from reprocessing old sky images from an asteroid hunt survey. The Near Earth Asteroid Tracking survey took thousands of photos that were searched once with older tools. Modern software can find fainter rocks in the same images.
Their new pipeline cleans and calibrates each image, pulls out light sources, and hunts for moving dots. It was run on ten nights of old photos as a first test. Finds are sent to the Minor Planet Center, the world clearing house for such sightings. The pipeline found fainter asteroids than the first pass did.
Catalog record
| CREATOR | C. R. Nugent, J. M. Bauer, O. Benitez, M. Blain, N. D’Souza, S. Garimella, M. Goldwater, Y. Kim, H. C. G. Larsen, T. Linder, K. Mackowiak, Z. McGinnis, E. Pan, C. C. Pedersen, P. Sadhwani, F. Spoto, N. J. Tan, P. Vereš, C. Xue |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-20032
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Reprocessing the NEAT Dataset: Preliminary Results Journal: PSJ astropy ( Astropy Collaboration et al. 2022 ; Astropy Collaboration et al. 2018 ; Astropy Collaboration et al. 2013 ) , astroquery ( Ginsburg et al. 2019 ) , photutils ( Bradley et al. 2023 ) , sklearn ( Pedregosa et al. 2011 ) , Astrometry.net ( Lang et al. 2010 ) , SkyBoT ( Berthier et al. 2006 ) , FindPOTATOS ( Nugent et al. 2025 ) C. R. Nugent, J. M. Bauer, O. Benitez, M. Blain, N. D’Souza, S. Garimella, M. Goldwater, Y. Kim, H. C. G. Larsen, T. Linder, K. Mackowiak, Z. McGinnis, E. Pan, C. C. Pedersen, P. Sadhwani, F. Spoto, N. J. Tan, P. Vereš, C. Xue Abstract We have created a new image analysis pipeline to reprocess images taken by the Near Earth Asteroid Tracking survey and have applied it to ten nights of observations. This work is the first large-scale reprocessing of images from an asteroid discovery survey in which thousands of archived images are re-calibrated, searched for minor planets, and resulting observations are reported to the Minor Planet Center. We describe the software used to extract, calibrate, and clean sources from the images, including specific techniques that accommodate the unique features of these archival images. This pipeline is able to find fainter asteroids than the original pipeline. I Introduction Asteroid discovery surveys image large swaths of the sky, discovering hundreds of near-Earth objects (NEOs) each month ( Jedicke et al. 2015 , e.g.) 1 1 1 Current monthly discovery rates can be found on the Minor Planet Center home page, https://minorplanetcenter.net/ . . Comparison of discovery statistics with population models ( Granvik et al. 2018 ; Nesvorný et al. 2024 , e.g.) indicates that together, surveys have made significant progress toward discovering the NEO population larger than 140 m in diameter. However, there is a portion of the NEO population that these surveys cannot detect on any given night: objects with locations relative to survey telescopes and the Sun that place them in the daytime sky. Over months or years, many NEOs drift out of this unobservable group and into the night sky, but some linger for decades. For example, consider the edge case 2002 AA29. Observations of this NEO were reported to the Minor Planet Center (MPC) from 2002 until 2004, after which it became unobservable. Because it has an orbital period of 0.99 years, it will remain unobservable from Earth until 2093. To extend observed arcs of NEOs that linger in the daytime sky, researchers can seek precovery observations in archival data ( Boattini et al. 2001 ; Gwyn et al. 2012 ; Vaduvescu et al. 2013 ; Perlbarg et al. 2023 ; Saifollahi et al. 2023 , e.g.) . Precovery observations can secure orbits, which is particularly valuable for objects that make close approaches to planets, including Earth. However, archival data can contain artifacts that are uncommon in modern data, requiring special care and ruling out the use of some software tools. Additionally, undocumented issues related to the images, such as telescope movement or timing errors, can result in erroneous results. Extracting quality data from archival images therefore requires significant time, and in some cases, new software development. We describe a new image processing pipeline applied to the archived Near Earth Asteroid Tracking (NEAT) survey, which operated from 1995 to 2007 and was led by Eleanor “Glo” Helin ( Helin & Dunbar 1990 ; Helin et al. 1997 ) . NEAT was a project of NASA’s Jet Propulsion Laboratory (JPL) and the Air Force Space Command. It employed several 1-meter class telescopes with CCD cameras and large fields of view. Together with Spacewatch ( Gehrels & McMillan 1982 ; Gehrels et al. 1986 ; Rabinowitz 1991 ) , NEAT was one of the first asteroid surveys to use digital imaging technology and search for asteroids using computers. It began surveying from Haleakala, Maui, HI, with the Ground-based Electro-Optical Deep Space Surveillance (GEODSS) camera. The telescope was remotely accessed from Pasadena, California. Data were sent from the telescope to the computing resources at JPL using a modem and a 1-800 telephone number with an effective transmission rate of 1.5 kbps. In 2001, the survey started operating at Palomar Observatory Oschin Schmidt Photographic Equatorial Telescope in Southern California using the TriCam camera. During its lifetime, NEAT discovered 41,227 minor planets and submitted over a million observations to the MPC. The GEODSS camera was initially paired with the U.S. Air Force one meter telescope; in 2000 it was moved to the Maui Space Surveillance Site 1.2 m telescope 2 2 2 Planetary Data System; https://pds.nasa.gov/ds-view/pds/viewContext.jsp?identifier=urn%3Anasa%3Apds%3Acontext%3Ainstrument%3Aamos.neat_maui_camera&version=1.1 . GEODSS had 4096\times 4096 pixels, with a 1.43 arcsecond/pixel scale on the first telescope and a 1.36 arcsecond/pixel scale on the second telescope. The CCD was read out by four separate amplifiers. Pravdo et al. 1999 reports a limiting magnitude of 19.1\pm 0.1 . Images were taken without filters. Inspection of a GEODSS image taken in 1995 shows a FWHM of 2.6 pixels. As its name implies, the TriCam camera consisted of three distinct 4096\times 4096 pixel CCD sets, or cameras. 3 3 3 Planetary Data System; https://pds.nasa.gov/ds-view/pds/viewContext.jsp?identifier=urn%3Anasa%3Apds%3Acontext%3Ainstrument%3Apalomar.oschin_schmidt_1m2.tricam&version=1.0 The cameras are arrayed vertically and distinguished via ‘a’, ‘b’ and ‘c’ labels. The survey took three exposures per field, the same camera was used for each exposure in a field. Each camera was read out by four amplifiers. The pixel scale for each camera is 1.01 arcseconds/pixel, therefore, each camera covers a bit more than a square degree. Images were taken without filters. The FWHM of sources is 3.7 pixels. A typical NEAT survey cadence is described in Pravdo et al. 1999 . However the archived data ( Bauer & Lawrence 2013 ) do not always match that described cadence. Here we provide details on the NEAT observations reprocessed in this work. These numbers may have changed over the course of the NEAT TriCam survey and cannot be assumed to apply to all NEAT TriCam observations. Between 120 and 462 fields were observed each night. Observing cadence information is available in the logs folder associated with each archived NEAT night in the Planetary Data System archive ( Bauer & Lawrence 2013 ) . According to those log files, observations generally had 60 s exposures with either 900 or 1800 s between exposures in the same field. However, some nights had 150 s exposures with 5400 s between exposures. Other nights consisted solely of 20 s exposures, either with 900 or 1800 s between exposures. The exposure time information in the PDS log files correlates with the exposure times recorded in the individual FITS file headers. However, the time between exposures as determined by the differences between the DATEOBS field of the FITS files and the times recorded in the PDS log files can be different by tens of minutes. We infer that the round number times recorded in the PDS log files were the planned cadence, and sometimes complicating real-world factors resulted in the executed cadence being different. In the decades since NEAT operated, there have been improvements in computing hardware and software. We have created a new data analysis pipeline to reprocess the NEAT dataset to find new