Hubble Extends Our Cosmic Horizon Back Through 97% of Cosmic History

Pascal Oesch,

In The Realm of Nebulae, Edwin Hubble wrote: “The history of astronomy is a history of receding horizons,” a quote which could not be more fitting to describe the story and the discoveries of the Hubble Space Telescope. During its 26 years in space, Hubble has steadily pushed our observational horizon to earlier and earlier cosmic times and transformed our view and understanding of how galaxies built up and evolved in the early universe. Starting from the Hubble Deep Field (Williams et al. 1996) and the discovery of z ∼ 4 galaxies from about 12 billion years in the past (Madau et al. 1996), Hubble has now found galaxies as far back as z ∼ 10, when the universe was only about 3% of its current age (e.g., Bouwens et al. 2011; Ellis et al. 2013; Oesch et al. 2014). This is a stunning achievement, and one that few expected to be possible at the time of Hubble’s launch.

The latest of Hubble’s accomplishments to expand our cosmic horizon of galaxies came as a surprise even for astronomers today. Using the powerful WFC3 infrared camera which was installed in 2009, Hubble discovered—and then spectroscopically confirmed (Oesch et al. 2016)—a very luminous galaxy at z = 11.1, only 400 million years after the Big Bang. This galaxy, named GN-z11, is the farthest object ever seen in the universe. It takes astronomers into a realm once thought only to be reachable with NASA’s upcoming James Webb Space Telescope. The new Hubble observations push back the frontier of galaxies that have accurately measured distances by a large margin: the previous record-holder was found at z = 8.68, more than 150 Myr after GN-z11 (Zitrin et al. 2015).

Figure 1: A zoom into the GOODS-North field, where the galaxy GN-z11 (shown in the inset) was detected based on images from the CANDELS survey (Grogin et al. 2011; Koekemoer et al. 2011). Using slitless spectroscopic observations with Hubble, the redshift of GN-z11 was measured to be z = 11.1, making this the most distant object ever discovered. The image shows the galaxy as it was 13.4 billion years in the past, just 400 million years after the Big Bang, when the universe was only 3% of its current age. Credit: NASA, ESA, P. Oesch (Yale University), G. Brammer (STScI), P. van Dokkum (Yale University), and G. Illingworth (University of California, Santa Cruz).

There is a fascinating story here. It turns out that the galaxy GN-z11 was seen in observations with WFC3’s IR predecessor, NICMOS, back in 2008. It was then thought to be too bright to really lie at such an extreme distance (Bouwens et al. 2010). When combined with the uncertain redshift from the noisy data, this NICMOS detection was not taken very seriously at that time. Astronomers continued to expect that the galaxy population at z ∼ 10–12 would be very faint, requiring very deep observations to be detectable. Surprisingly, however, most of the z > 9 galaxies currently known were not discovered in ultra-deep imaging such as the Hubble Ultra-Deep Field, but in shallower surveys covering wider areas. GN-z11 was identified among a sample of four luminous z > 9 galaxy candidates in observations with the WFC3/IR camera in the GOODS-North field as part of the public CANDELS survey (Oesch et al. 2014).

The new images with WFC3/IR convinced astronomers that GN-z11 was highly likely to be at very early cosmic times. Nonetheless, proof in the form of a spectroscopic redshift measurement was required. This was the goal of Hubble program GO-13871 (PI: Oesch). While previous distance measurements for early galaxies mostly relied on the identification of the redshifted Lyα emission line with ground-based 10m-class telescopes, it has recently been found that these Lyα emission lines disappear rapidly for galaxies beyond z > 6 (e.g., Schenker et al. 2012; Treu et al. 2013). This can be explained by the increasingly neutral intergalactic medium at these early cosmic times, which scatters and absorbs the Lyα photons.

However, Hubble does not require a Lyα emission line for a redshift measurement. Orbiting outside of Earth’s atmosphere, a key advantage of Hubble is that it is subject to much lower background radiation in the near infrared compared to ground-based telescopes. The use of grism spectroscopy with the very sensitive WFC3/IR camera allowed astronomers to successfully detect the continuum of GN-z11 (at 26th magnitude). In particular, a clear break was measured at 1.47 micron, which indicates a redshift of z = 11.1, even higher than the original estimate based purely on the photometry. This redshift is very close to the limit of what Hubble will ever be able to see, as galaxies even further away will shift out of the wavelength range covered by Hubble’s instruments.

Figure 2: The measurement of GN-z11 pushes our cosmic horizon close to the formation of the first galaxies. This discovery provides a very tantalizing preview of many even more distant galaxies Webb will be able to detect and characterize by the end of this decade. The graphic shows a timeline of the universe, stretching from the present day (left) to the Big Bang (right). Credit: NASA, ESA, B. Robertson (University of California, Santa Cruz), and A. Feild (STScI).

The discovery of GN-z11 at z = 11.1 in the current Hubble images challenges our theoretical understanding of galaxy formation in the early universe. The latest models have all predicted that existing Hubble surveys would be too small to detect such a luminous galaxy, and such a detection would require a search area 10–100 times larger (e.g., Mashian et al. 2015; Trac et al. 2015; Mason et al. 2015; Waters et al. 2016). GN-z11 may thus indicate something fundamentally new about how efficiently early galaxies formed out of the primordial gas after the Big Bang. Larger area surveys are now needed to more accurately measure the cosmic abundance of galaxies as bright as GN-z11 in the early universe. This can already be achieved with a substantial investment of Hubble time, and would be a striking legacy for Hubble. It will certainly be done with NASA’s planned Wide-Field Infrared Survey Telescope (WFIRST), which will have the ability to find thousands of similarly bright and distant galaxies.

These findings also have exciting consequences for what astronomers will be able to discover with Webb after its launch in 2018. The confirmation of GN-z11 proves that galaxy build-up was well underway at 400 Myr after the Big Bang. GN-z11 was detected in images taken with the Spitzer Space Telescope, which indicates that it had already built about a billion solar masses in stars. This build-up must have started very soon after the Big Bang.

With Hubble’s discovery and spectroscopic confirmation of GN-z11, we have now explored 97% of cosmic history. While Hubble is reaching its limits in further extending our cosmic horizon, Webb will soon take over and take us back to a time when galaxies were first forming. The progenitors of GN-z11 will be easy targets for Webb well beyond z = 11 and, true to the statement of Edwin Hubble, the history of astronomy will continue to be rewritten.


Bouwens, R. J., et al. 2010, ApJ, 725, 1587

Bouwens, R. J., et al. 2011, Nature, 469, 504

Ellis, R. S., et al. 2013, ApJL, 763, 7

Grogin, N. A., et al. 2011, ApJS, 197, 35

Koekemoer, A. M., et al. 2011, ApJS, 197, 36

Madau, P., et al. 1996, MNRAS, 283, 1388

Mashian, N., et al. 2016, MNRAS, 455, 2101

Mason, C. A., et al. 2015, ApJ, 813, 21

Oesch, P. A., et al. 2014, ApJ, 786, 108

Oesch, P. A., et al. 2016, ApJ, 819, 129

Schenker, M. A., et al. 2012, ApJ, 744, 179

Trac, H., et al. 2015, ApJ, 813, 54

Treu, T., et al. 2013, ApJL, 775, 29

Waters, D., et al. 2016, arXiv:1604.00413

Williams, R. E., et al. 1996, AJ, 112, 1335

Zitrin, A., et al. 2015, ApJL, 810, 12