Advanced gravitational waves detector CALibration for accurate COsmology – ACALCO
Advanced gravitational waves detector CALibration for accurate COsmology
Improving the Gravitational Wave detector calibration accurary
Reaching a sub-percent accuracy of the calibration signal to match the detector sensitivity improvement.
After the first observation of gravitational waves (GW) in 2015, the LIGO and Virgo collaborations made in 2017 the first observation of GW coming from the final cycles of two orbiting neutron stars that eventually collided. The major breakthrough with this event (GW170817) was the localization of the source by GW detectors, that enabled the detection of an optical counterpart. With the simultaneous observation of a gamma ray burst, this was the start of GW multimessenger astronomy. With this single event, it was possible to make a new independent measurement of the Hubble constant (H0), a key parameter in cosmology, for which there is currently a “tension”: a significant difference of about 8%, between the measurements using nearby sources (Supernovae) or early-universe information (CMB). This single event has demonstrated the potential of the H0 measurement with GW events. Since the first event, we have witnessed an acceleration of the number of detections during each LIGO-Virgo data taking. This large number of events should decrease the statistical error of the GW H0 measurement, to a level low enough to bring meaningful input to the H0 understanding. But unfortunately, if nothing is changed, the systematic uncertainty coming from the absolute calibration of the GW detectors will dominate and limit this measurement. Indeed, GW detectors, even if they are very sensitive instruments, are not very accurate. Their absolute calibration uncertainty at the start of the project was in the 2 to 5% range. This uncertainty propagates directly to the H0 measurement, and must be reduced to extract the best result from the O5 observing run and future data taking runs. Let’s notice that the correct localization of a source in the sky, another key part for the discovery of the faint optical transients and therefore the direct GW H0 measurement, relies on the measurement of the relative timing and amplitude of the signals seen by the GW detectors acting as a network. Here again, having an accurate calibration, especially cross-calibration of the GW detectors is another goal to pursue and would benefit from an accurate absolute calibration of each GW detector. But the measurement of the GW H0 constant is not the only result directly impacted by calibration uncertainties. For instance, the two following results would benefit from calibration improvements: 1) Measuring population rates: The typical current 5% distance uncertainty translate to a 15% volume and therefore to a rate uncertainty which is not so negligible when computing a rate with about 50 BBH events, as observed during O3. 2) Tests of general relativity: these tests check that amplitude and phase evolutions of the observed GW signal, when sweeping through the frequency band, are consistent with the waveform predicted by general relativity. Therefore, an unaccounted frequency dependence of the calibration in amplitude or phase will bias these tests.
GW detectors are measuring space-time deformations from GW by comparing the length of the two arms of large interferometers. Usual GW calibration methods are moving one of the mirrors by a well-known amount and calibrating accordingly the detector output.
The reference method at the proposal time, known as PCal (for Photon Calibrator), consists in using the radiation pressure of an auxiliary laser whose power is modulated to introduce the expected displacement [PCal-O3]. This is a versatile solution, but it is a solution which is facing strong challenges when measuring the absolute power to a fraction of percent.
A new method based on the generation of a variable Newtonian gravitational field produced by rotating masses (Newtonian Calibrator or NCal) has been explored [NCal-O2, NCal-O3]. This has the main benefit of replacing the power measurement by metrology measurements, that should reach sub-percent accuracy. However, the maximum frequency of the signal is limited by mechanical constrains, the maximum rotation speed of a rotor.
The calibration effort of the Virgo detector is carried on by the two labs participating to the ACALCO project. The LAPP team is expert on PCals, while the IPHC team is focusing on NCals. The systematic uncertainty reached at the start of the project was 1.4% for both methods, with signal injected up to 120 Hz for the NCal and several kHz for the PCal. The main effort benefit of the ACALCO grant was to provide support to improve these -uncertainties the for the Virgo O4 data taking (2024-2025).
For the PCal system, the main effort was to improve the power measurement of the laser beam. More specifically, the effort was on the inter-calibration with LIGO, which requires dedicated measurement at one of the LIGO sites, exchange of reference power-meters with LIGO and metrology institutes, the development of optical bench to intercalibrate power integrating spheres, and multiples measurement to assess the stability over time. Together with other improvements, this reduced the systematic uncertainty down to 0.5 % for frequencies up to 1 kHz, with the uncertainty increasing at higher frequencies.
The Virgo NCal system built for O4 was brand new. A total of six rotors were installed around one of the end mirrors in 2023. After an extensive commissioning period, leading to adjustments like the change of material used for the rotors, the achieved uncertainty at the start of the O4 run was 0.17%. The reliability of the system allows for continue operations, injecting a set of permanent “calibration lines” in the Virgo interferometer. The improvement of the frequency band was also demonstrated with test signals injected up to 140 Hz. During the ACALCO project, the NCal developments focus on providing an accurate and reliable signal, to be used 24/7 during a multi-year data taking, as the O4 run. Therefore, a modest operation frequency was privileged (a signal at 36 Hz in h(t)). This was successful since several NCal were operating at Virgo during the entire Virgo O4 run without mechanical failures. The multiple NCals, located on opposite mirror sides, played an important role in reducing the system uncertainty associated to the mirror to NCal distance. Another key development was the reduction of the parasitic coupling, especially the magnetic coupling, which require a change of the material (from Aluminum to PVC and later Peek) used for the NCal rotors. Having a dedicated PhD student to carry on all the needed investigations was essential for the success of the project.
The low uncertainty of the NCal system made it the reference system for the absolute calibration of Virgo for the O4 run. The PCal system was therefore recalibrated on the NCal system just before the start of O4, providing accurate reference signal on a large frequency band, thanks to the flexibility of the PCal. This demonstrated the benefits of using these two complementary techniques.
Thanks to the ACALCO grant, the accuracy of the calibration signals using by Virgo during the O4 run has been improved by almost an order of magnitude. The Virgo data were used to provide public alerts for GW astrophysical event and will be released after an 18 months period.
While the ACALCO project focused on the improvement of the accuracy at low frequency, the enlargement of the frequency range (which is challenging in term of reliability for the NCal) remains a key topic to be investigated.
The first new measurement of the Hubble constant (H0) using gravitational waves has been made in 2017. The upcoming data taking of the upgraded LIGO and Virgo detectors should bring results of astrophysical relevance for H0, at the condition that significant progresses are made with the detectors calibration. The ACALCO project objective is to do this R&D to reach a sub-percent absolute calibration uncertainty of the Virgo detector by 2025. The plan is to pursue a combined approach. On the one hand, develop a new calibration system based on the generation of a variable Newtonian gravitational field produced by rotating masses. This technique has been recently explored and could become the new absolute calibration reference. On the other hand, improve the technique which uses the radiation pressure of an auxiliary laser whose power is modulated, to use it as an accurate cross reference and as a system to extend the calibration the whole Virgo detector frequency band.
Project coordination
Benoit Mours (Institut Pluridisciplinaire Hubert Curien - IPHC (UMR 7178))
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
Partnership
IPHC Institut Pluridisciplinaire Hubert Curien - IPHC (UMR 7178)
LAPP LABORATOIRE D'ANNECY DE PHYSIQUE DES PARTICULES
Help of the ANR 425,600 euros
Beginning and duration of the scientific project:
- 48 Months
