A Search for Technosignatures Around 11,680 Stars with the Green Bank Telescope at 1.15-1.73 GHz

Brief

The UCLA SETI team used the 100 m GBT (1.15–1.73 GHz) to observe 62 TESS Objects of Interest, capturing emissions from ~11,680 stars over 2020–2023. Their pipeline detected 41.2 million narrowband signals; 99.43% were auto-rejected as RFI and all ~500 surviving candidates were confirmed anthropogenic on visual inspection, extending an 8-year record of 82 million detections with zero follow-up-worthy signals. A signal-injection analysis showed the UCLA pipeline recovers 94.0% of injected chirp signals (98.7% outside dense RFI bands), versus only 5.7% for a Breakthrough Listen-style incoherent-averaging pipeline, a difference that materially affects transmitter-prevalence calculations. The resulting upper limits are: at 95% confidence, fewer than 6.6% of stars within 100 pc host a transmitter with EIRP > 10^13 W, and fewer than 3×10^-4 of stars within 20,000 ly host one with EIRP > 5×10^16 W.

Metadata

Category

Search

Venue

The Astronomical Journal

Type

Peer-reviewed

Year

2023

Authors

Jean-Luc Margot, Megan G. Li, Pavlo Pinchuk, Nathan Myhrvold, Larry Lesyna

DOI

10.3847/1538-3881/acfda4

arXiv

2308.02712

Access

Open access

Length

831.4 K

Programs

UCLA SETI, Breakthrough Listen, Project Cyclops

Instruments

Green Bank Telescope (GBT) L-band receiver 1.15–1.73 GHz, VEGAS baseband recording backend

Data sources

Gaia DR3 (Gaia Collaboration 2023), TESS Objects of Interest (TOI catalog), NASA Exoplanet Archive, Breakthrough Listen HSR power spectra (Ma et al. 2023 ML candidates)

Tags

SETI, technosignature, radio-astronomy, narrowband-search, pipeline-analysis, transmitter-prevalence

Key points

• Sample: 11,680 stars observed in 62 GBT beam pointings aligned with TESS Objects of Interest; median target distance 2,288 pc (7,461 ly), maximum 12,664 pc (41,305 ly).p.3
• Total detections across 8 years of UCLA SETI: 82 million narrowband signals; not one has merited follow-up observations.p.6
• UCLA pipeline end-to-end efficiency: 94.0% of 10,000 injected chirp signals recovered; 98.7% when dense-RFI frequency ranges are excluded. BL-like incoherent pipeline recovered only 5.7% of the same injections, roughly 16× worse.p.7
• The BL-like pipeline's poor recovery stems from Doppler smearing: 51-spectrum incoherent averaging extends integration to ~17 s, causing drift rates above ±0.15 Hz/s to smear across frequency bins; only 99.1% of BL-recovered signals had drift rates within ±1 Hz/s.p.7
• 95% confidence upper limit: fewer than 6.6% of stars within 100 pc host a transmitter detectable with EIRP > 10^13 W; for the 20,000 ly volume, the fraction is at most 3×10^-4 at EIRP > 5×10^16 W.p.1
• Sensitivity benchmark: an Arecibo Planetary Radar equivalent (EIRP = 2.2×10^13 W) is detectable at 415 ly; 1,000 Arecibos at 13,123 ly; transmitters at the galactic center would require 4,130 Arecibo equivalents.p.5
• Of the 8 ML-identified candidate signals from Ma et al. (2023), the UCLA pipeline natively detected 5 (MLc3–5, MLc7–8) without invoking ML; all were subsequently identified as RFI via presence in OFF scans.p.7
• Narrowband technosignature rationale: natural astrophysical masers (narrowest reported OH maser: 550 Hz) are roughly two orders of magnitude wider than the proposed <10 Hz threshold, making any sub-10 Hz detection unambiguously non-natural.p.2

Verbatim

• “It is remarkable that, in over 82 million narrowband signal detections obtained during an 8-year period (Table 1), not a single signal has merited follow-up observations.”

  p.6

• “Based on our observations, we can state at the 95% confidence level that fewer than 6.6% of stars within 100 pc host a transmitter that is detectable in our search (EIRP > 10 13 W). For stars within 20,000 ly, the fraction of stars with detectable transmitters (EIRP > 5 × 10 16 W) is at most 3 × 10 − 4 .”

  p.1

• “The UCLA SETI pipeline recovers 94 . 0% of the injected signals over the usable frequency range of the receiver and 98 . 7% of the injections when regions of dense radio frequency interference are excluded.”

  p.1

• “An example of such a technosignature is a narrowband (say, < 10 Hz at gigahertz frequencies) signal from an emitter located beyond the solar system. Detection of a signal with these characteristics would provide sufficient evidence for the existence of another civilization because natural settings cannot generate such signals.”

  p.2

• “With these parameters, an Arecibo Planetary Radar (EIRP=2 . 2 × 10 13 W) is detectable at 415 ly and a thousand Arecibos can be detected at 13,123 ly.”

  p.5

Most interesting

• The Breakthrough Listen-style incoherent pipeline is effectively blind to signals drifting faster than ±0.15 Hz/s, yet the UCLA pipeline's drift-rate ceiling of ±8.86 Hz/s is needed to cover ~73% of confirmed exoplanet orbital accelerations.
• A single GBT beam at 1.42 GHz (8.4 arcmin) sweeps up 11,680 stars per survey field, most of them serendipitous background targets rather than the 62 primary TESS Objects of Interest.
• Citizen scientists on arewealone.earth are inspecting tens of thousands of the ~230,000 signals that survived automated rejection, an embedded crowd-sourced vetting layer within a professional survey.
• The probability of an accidental frequency+drift-rate match in the injection-recovery analysis is less than one in a billion, establishing the injection test as a high-confidence pipeline audit.
• Over the full 8-year UCLA SETI program, 55,555 stars have been observed; the cumulative Modified Drake Figure of Merit reached 3.37×10^33 Hz²·m³·W^(-3/2).
• The narrowest known OH maser line (550 Hz at 1612 MHz) is roughly 55× wider than the <10 Hz bandwidth threshold that would render a radio signal unambiguously non-astrophysical.