Cosmic magnifying lenses distort view of distant galaxies EMBARGOED UNTIL: 1 PM EDT / 11 A.M. ARIZONA TIME, January 12, 2010 Looking deep into space, and literally peering back in time, is like experiencing the universe in a house of mirrors where everything is distorted through a phenomenon called gravitational lensing. Gravitational lensing occurs when light from a distant object is distorted by a massive object that is in the foreground. Astronomers have started to apply this concept in a new way to determine the number of very distant galaxies and to measure dark matter in the universe. Though recent progress has been made in extending the use of gravitational lensing, a letter published in Nature on Jan. 13 makes the case that the tool may be even more necessary than originally thought when looking at distant galaxies. Albert Einstein showed that gravity will cause light to bend. The effect is normally extremely small, but when light passes close to a very massive object such as a massive galaxy, a galaxy cluster, or a supermassive black hole, the bending of the light rays becomes more easily noticeable. When light from a very distant object passes a galaxy much closer to us, it can detour around the foreground object. Typically, the light bends around the object in one of two, or four different routes, thus magnifying the light from the more distant galaxy directly behind it. This natural telescope, called a gravitational lens, provides a larger and brighter --- though also distorted --- view of the distant galaxy. These distortions, which stretch beyond the limits of the Hubble Space Telescope, can be effectively handled by a new space telescope on the drawing boards --- the James Webb Space Telescope (JWST). A very massive object --- or collection of objects --- distorts the view of faint objects beyond it so much that the distant images are smeared into multiple arc-shaped images around the foreground object. According to Rogier Windhorst, one of the letter's authors and a professor at the School of Earth and Space Exploration in Arizona State University's College of Liberal Arts and Sciences, this effect is analogous to looking through a glass coke bottle at a light on a balcony and noticing how it is distorted as it passes through the bottle. Cosmologists such as Windhorst believe that gravitational lensing likely distorted the measurements of the flux and number density of the most distant galaxies seen in the recent deep near-IR surveys with the Hubble Space Telescope Wide Field Camera 3. When you look back to when the universe was young, you are seeing extremely early objects (also known as ``First Light'' objects) that are very far away. The older and farther away the object, the more foreground universe there is to look through, which means the greater the chance that there will be something heavy in the foreground to distort the background image. This research suggests that gravitational lensing is likely to dominate the observed properties of very early galaxies, those that are at most 650-480 million years old (now seen with Hubble at redshifts of z > 8-10, respectively). The halos of foreground galaxies when the universe was in its heydays of star formation (about 3-6 billion years old and at a lower redshift of z=1-2) will gravitationally distort many of these very early objects. ``We show that gravitational lensing by foreground galaxies will lead to a higher number of galaxies to be counted at redshifts z>8-10. This number may be boosted significantly, by as much as an order of magnitude. If there existed only three galaxies above the detection threshold at redshifts z>10 in the Hubble field-of-view without the presence of lensing, the bias from gravitational lensing may make as many as 10-30 of them visible in the Hubble images,'' explains Windhorst. ``In this sense, the very distant universe is like a house of mirrors that you visit at the State Fair --- there may be fewer direct lines-of-sight to a very distant object, and their images may reach us more often via a gravitationally-bent path. What you see is not what you've got!'' Future surveys will need to be designed to account for a significant gravitational lensing bias in high-redshift galaxy samples. Only the JWST --- if it gets finished as designed --- can ultimately make sense out of this gravitationally biased distant universe, because it will have exquisite resolution and sensitivity at longer wavelengths to disentangle these very distant objects from the foreground lensing galaxies. This work is too hard to do with Hubble's Wide Field Camera 3 at redshifts z>10, because at Hubble's resolution one literally can no longer see the forest for the trees at these extreme distances. ``The very distant Universe is literally throwing us an enormous curve-ball here that very few saw coming'', says Windhorst. ``Only the Webb telescope will have the batting average to properly field this one.'' ``Our suggestion of a significant gravitational lensing bias in very distant galaxy samples is crucial to design surveys of the first galaxies, which is a central part of JWST's mission,'' says Windhorst. ``This work clearly shows that we now need the superb resolution and sensitivity of JWST more than ever to disentangle the First Light forest from the foreground trees. We will also need a next generation of object finding algorithms, since the current software is simply not designed to find these rare background objects behind such dense foregrounds. It's like finding a few needles in the mother-of-all-haystacks, while Mother Gravity also hides those needles in those parts of the haystack where the hay is densest! '' ### The Nature Letter by J. Stuart B. Wyithe (Univ. of Melbourne, Australia), Haojing Yan (Ohio State Univ.), Rogier A. Windhorst (Arizona State Univ.), and Shude Mao (Jodrell Bank, Manchester University, UK) appeared in Vol. 469 on Thursday Jan. 13, 2011. Caption: This color composite image of the Hubble Ultra Deep Field was created using data taken by its WFC3 and ACS instruments. The green circles mark the locations of candidate galaxies at redshifts of about z = 8--10, where the Universe was only 650--480 million years old. About 20-30% of these candidates are very close to foreground galaxies, which is consistent with the prediction that a significant fraction of galaxies that we could see at these very high redshifts are gravitationally lensed by individual foreground galaxies. Photo credit: NASA, ESA, Z. Levay and A. Feild (STScI). Data Credit: S. Beckwith (STScI) and the HUDF team; G. Illingworth and R. Bouwens (University of California, Santa Cruz), and the HUDF09 Team. ASU SOURCES: Rogier Windhorst, rogier.windhorst@asu.edu 480.540.0816 (c) or 480.965.7143 (o) MEDIA CONTACT: Nicole (Staab) Cassis, nstaab@asu.edu 602.710.7169 Arizona State University (www.asu.edu) School of Earth and Space Exploration (http://sese.asu.edu) Tempe, Arizona, 85287-1404, USA ======================================================================== All materials related to this press release are also kept on these URL's: 1) http://hubblesite.org/newscenter/archive/releases/2011/04 2) All files related to this press release are also on this ASU web-site: http://www.asu.edu/clas/hst/www/nature11/ Pressrel*.txt --- Press releases by Nature, STScI, ASU, OSU, Melbourne NatureWyitheProofs.pdf --- Wyithe et al. 2011 Nature Letter (Vol. 469, p. 1-4) NatureSupplementary.pdf --- Wyithe et al. Nature Supplementary information. AAS2011Poster*pdf --- Windhorst et al. AAS poster 347.09 describing the Wyithe et al. result in context of the James Webb Space Telescope. Cartoon3d-planes-m.jpg --- Describes how gravitational lensing bias works. Nature_FIG0s_press.jpg --- More technical cartoon describing lensing bias. HudfBVizYJH_NC* --- Best available 506 orbit HUDF image in 7 filters (BVizYJH) from 0.4--1.7 microns, nearly true color, log-stretch. HudfBVizYJH* --- Best available 506 orbit HUDF image in 7 filters (BVizYJH). Green circles are Y-band dropouts (z=8 candidates); Red circles J-band dropouts (z>10 candidates). These candidates have yet to be confirmed by spectra. s1104a-hudf-m.jpg --- Smaller version of the HUDF color image with candidates. hudfBVizYJHA057A066.jpg --- Full color example of three z=8 candidates (green circles) that suggest gravitational lensing bias. P1104-hudf-m.jpg --- Other examples of possible Gravitational Lensing bias. figures/fig_*drop-A*.jpg --- [in subdirectory figures/ ] other examples of possible gravitational lensing bias. Yan_AAS2011_press.ppt --- H. Yan's 2011 AAS press presentation on lensing bias. Windhorst_AAS2011_press.pdf --- Windhorst's presentation on implications for JWST 2) 1) The Java tool AHaH that let's you zoom through the Hubble UltraDeep Field (HUDF) fully three dimensionally with proper relativistic geometry: http://www.asu.edu/clas/hst/www/jwst/HUDFjavatool/index.html http://www.asu.edu/clas/hst/www/jwst/HUDFjavatool/download.html [AHaH stands for ``Appreciating Hubble at Hyperspeed''. This Java tool does not yet include Gravitational Lensing bias, but stay tuned ... ] 3) The HUDF clickable map, that allows the user to get redshift estimates of most HUDF objects that one can click on. It's full resolution version is available on: http://www.asu.edu/clas/hst/www/jwst/clickonHUDF/index.html