In:
Science, American Association for the Advancement of Science (AAAS), Vol. 380, No. 6649 ( 2023-06-09)
Kurzfassung:
The Hubble constant ( H 0 ) quantifies the rate at which the Universe is expanding. It has been predicted that observations of a supernova (SN) that is multiply imaged by a foreground gravitational lens could, in principle, be used to measure H 0 . The time delays between the SN’s multiple images are expected to be inversely proportional to the value of H 0 . In late 2014, a SN was found that was strongly lensed by the galaxy cluster MACS J1149.5+2223. Dubbed SN Refsdal, it appeared in four images around a galaxy that was a member of the cluster. Models of the gravitational lens predicted that a fifth image of SN Refsdal would appear within the next year. Multiple teams made blind predictions for the timing and relative brightness of the SN’s reappearance. RATIONALE The two most precise previous measurements of the Hubble constant are inconsistent with each other (at 〉 5σ significance), which is known as the Hubble tension. Measurements of thermonuclear supernovae (SNe) in the nearby Universe have yielded H 0 = 73.0 ± 1.0 km s − 1 Mpc − 1 . Observations of the cosmic microwave background (CMB) have led to a lower value of H 0 = 67.4 ± 0.6 km s − 1 Mpc − 1 , assuming standard cosmology. If this tension is confirmed using independent measurements, it could indicate problems with standard cosmology. An alternative cosmology, such as those involving an additional early dark energy or more species of particles, might be required to reconcile the measurements. The predicted additional image of SN Refsdal appeared in 2015. Follow-up observations have allowed the time delays and magnification ratios between the multiple images to be determined. We combine those measurements with the pre-reappearance model predictions to perform a blinded calculation of the value of H 0 . RESULTS We compute the likelihood of the time delay, magnification, and position measurements given each lens model’s predictions after rescaling the time delay predictions for different values of H 0 . This approach weights each model’s contributions in a combined constraint on H 0 . Weights are assigned according to each model’s ability to reproduce the observations that are independent of H 0 (location of the reappearance, magnification ratios, and ratios of time delays). We perform two estimates of H 0 : (i) from all lens models that made published predictions before the reappearance was observed and (ii) using a subset of those models, by selecting those that are most consistent with the observations and dominate the weights. CONCLUSION We infer a value of H 0 of 64.8 − 4.3 + 4.4 km s − 1 Mpc − 1 using the full set of eight pre-reappearance models and of 66.6 − 3.3 + 4.1 km s − 1 Mpc − 1 from the two preferred models. Our results are most consistent with the H 0 value measured from the CMB but do not exclude the higher value from nearby SNe. We used a simulation of a galaxy cluster lens to verify that the uncertainty on our measurement of H 0 is consistent with expectations. The ability of the lens models to reproduce the positions of the SN images also implies an expected uncertainty on H 0 , which we find agrees with our constraints. The best agreement between lens models and observations that are independent of H 0 is achieved by the models that were constructed by assigning dark-matter halos to both the cluster and to individual galaxies in the cluster. Appearance and reappearance of SN Refsdal, strongly lensed by the galaxy cluster MACS J1149.5+2223. Both panels were taken with the Hubble Space Telescope. ( A ) Images S1 to S4 (labeled) of SN Refsdal appear in an Einstein cross configuration in 2014. Models predicted the appearance of an additional image SX in the region denoted by the red box. ( B ) Observation of SX in spring 2016 after its appearance in late 2015. In (A), colors were assigned to coadded images taken in these filters: blue, F606W ( V -band) and F814W ( I -band); green, F105W ( Y -band) and F125W ( J -band); and red, F140W (between the J- and H -bands) and F160W ( H -band). The image in (B) used only the F125W filter, so it is shown in grayscale.
Materialart:
Online-Ressource
ISSN:
0036-8075
,
1095-9203
DOI:
10.1126/science.abh1322
Sprache:
Englisch
Verlag:
American Association for the Advancement of Science (AAAS)
Publikationsdatum:
2023
ZDB Id:
128410-1
ZDB Id:
2066996-3
ZDB Id:
2060783-0
SSG:
11
