2. Science with HST10X
2.1. Introduction

The Hubble Space Telescope is arguably one of the most important and successful scientific endeavors undertaken in the twentieth century. Hubble, a modest-sized 2.4-m telescope, outperforms much larger terrestrial telescopes because it is diffraction limited, and because the sky seen from orbit is darker than the terrestrial night sky. If we increase the diameter of Hubble to 8.4-m, a diameter comparable to Keck and the VLT, the increase in capability will be comparable to that which was first achieved by Hubble's launch and subsequent repair.

The Science Team was asked to define the science that might be enabled by a ten-fold enhancement to the optics of the Hubble Space Telescope (HST10X). We assume an 8.4-m aperture, afocal telescope in front of the existing HST optics. The science enabled by such an enhancement is only beginning to be considered but will certainly include:

DETECTION OF EARTH-LIKE PLANETS ORBITING NEARBY STARS at distances up to 10 parsecs. Furthermore, HST10X will enable spectroscopic examination of Earth-like planets to search for atmospheric oxygen, a certain sign of life, out to about 5 parsecs. The best wavelength to look for Earth-like planets is between 0.6 & 0.8 micrometers which is right in the middle of the HST bandpass.

Obtain additional Hubble Deep-Field (HDF) observations AT 15 PER DAY instead of the 150 orbits in 10 days for the original, at five times the angular resolution. A one-orbit F814W HST10X exposure will be deeper than the original HDF and have 14 times more spatial information per galaxy. HST10X will allow measurement of the spectra and spectra-energy distributions of high-redshift galaxies that cannot be reached with the Keck and VLT telescopes.

The "Key Projects" of 100-1,000 orbits and the "Legacy Projects" of 500-1,000 orbits proposed for HST's second decade could be done in days instead of months.

Confirm, or deny, the HDF implication that we have already seen into and beyond the era when galaxies first formed.

HST10X will extend HST's extraordinary capability for measuring Cepheid distances from the present 20 Mpc to 100 Mpc, thereby determining H0 (averaged over 100 Mpc) in one step, and thus avoiding the systematic errors introduced by secondary distance indicators. The Cepheid distances will map large-scale flows within 100 Mpc and delineate the large-scale distribution of dark matter that drives the flows.

HST10Xís large increase in spatial resolution (3.5 to 6.5) and spatial information (12 to 40) will capitalize on HSTís stunning discovery of proto-planetary disks in the Orion nebula to study the formation of stellar and planetary systems. HST10X may actually be able to detect the signature of planets forming in the proto-planetary disks.

By using faint quasars as UV background sources, HST10X could make tomographic maps of the full "cosmic web" of the filamentary distributions of hot (shocked) and warm (photoionized) baryons left over from the epoch of large-scale structure formation.

In section 2.2 we quantify the performance of HST10X, and compare that performance to the performance of present and planned HST instruments. In section 2.3 we give examples of outstanding science programs that can be undertaken with HST10X.

2.2. HST10X Performance
2.2.1 Optical Performance

Performance estimates for HST10X are based on the assumptions in Table 2.1.

When deep exposures span many orbits, as in the Hubble Deep Field, stepping the telescope a fraction of a pixel in x and in y between exposures and then properly combining the stepped images can improve the resolution of under-sampled images. This process, referred to as "dithering", yields a resolution that is close to one pixel in under-sampled images. Even with dithering, structure is difficult to resolve in high redshift galaxies. This is illustrated in Figure 2.1., a postage stamp extraction from the HDF F814W composite image. The galaxy in the center has a measured redshift of z = 2.26. The maximum dimension of the galaxy is ~0.3 arcsec.

Figure 2.2. shows a simulation that compares the resolution of a HST WFPC2 dithered image with an HST10X exposure of the antenna galaxy. The image with WFPC2 resolution shows little more than the fact that the galaxy is asymmetrical, and has color gradients from the center to edge. In contrast, the simulated HST10X image shows the merger of two nucleated galaxies, and reveals dust and bright knots of star formation throughout the two merging galaxies. The original image was a true-color composite of four WFPC2 images of the "Antennae Galaxies." The original image was scaled to give 0.3" separation between the nuclei of the two interacting galaxies. This angular size is the characteristic size of high redshift galaxies in the HDF. The scaled image was then resampled to the resolution of the WFPC2, Figure 2.2. left, and to the resolution of HST10X, Figure 2.2. right. The simulation illustrates the large gain in spatial information achieved, ~14 times, by increasing the resolution by a factor of ~3.7 (FWHM = 0.14" in F606W images).
 
 

Figure 2.2. Simulated Images of Interacting Galaxies at High redshift
WFPC2 resolution	HST10X Resolution

Table 2.3. compares the areal coverage of HST10X+ACS with the areal coverage of the WFPC2 and the ACS. The areal coverage of the ACS Wide Field Camera (ACS) would be 2.5 times the areal coverage of the present PC2.

Table 2.3. HST10X Areal Coverage (Square Arc-minutes) WFPC2 ACS HST10X + ACS WF2 PC2 WFC HRC WFC HRC 5.33 0.38 11.65 0.18 0.95 0.015

In Table 2.4. the stellar limiting magnitudes of the WF2, ACS, and HST10X+ACS are compared. The calculations assume that one orbit is divided into two 1200s exposures, the star is a black body with an effective temperature of 10,000 K, and the sky brightness at 57 degrees ecliptic latitude (the latitude of the HDF). In the cosmologically important F814W filter, HST10X+ACS reaches 4 magnitudes deeper than the present WF2 camera, and 2.4 magnitudes deeper than the ACS without the HST10X enhancement.

Table 2.4. HST10X Limiting Magnitudes (SNR = 10) Filter WF2 ACS WFC HST10X +ACS WFC F555W 26.55 27.8 30.2 F814W 25.70 27.4 29.8

Table 2.5. shows the number of ACS and WF2 Orbits required to equal the stellar limiting magnitude of one HST10X Orbit. It shows that it is effectively impossible for WF2 to reach the same stellar limiting magnitude as HST10X by summing orbits, and impractical for ACS in all but HDF-type observations.

Table 2.5. ACS/WF2 Orbits to Equal One HST10X Orbit Filter ACS WF2 F555W 35 410 F814W 47 1000

In the limiting case of a uniform surface brightness galaxy with IGal<< ISky, and an exposure that is sky-noise limited, the signal-to-noise ratio summed over the galaxy is given by:

2.2.2. UV Performance

The throughput enhancements to the HST-STIS/COS UV spectroscopic capabilities afforded by the HST10X enhancement will surpass important thresholds that enable new science. Assuming Al+MgF2 coatings on the new primary and secondary mirrors with UV reflectivities similar to those achieved on the STIS flight optics (~80-85%), the throughputs of the HST spectrographs will be increased by more than a factor of 8, bringing targets ~2.3 magnitudes fainter within reach. Approximate limiting sensitivities for the various spectroscopic modes to achieve S/N = 10 per spectral resolution element in 40,000s at 1400 Å are given in Table 2.6.

The HST10X enhancement will open the possibility of direct imaging of planetary systems of nearby stars. Most exciting will be the potential for discovery by direct detection and subsequent analysis of Earth-like planets in the nearest systems. If this potential is realized, the telescope is likely to be the first to find terrestrial planets, with sensitivity to systems out to 10 parsecs, and would be able to make spectroscopic analysis for water and oxygen in terrestrial planets found within about 5 parsecs. It complements TPF, which will be sensitive to the thermal emission and spectra of more distant terrestrial planets.

The best wavelength to look for Earth-like planets is between 0.6 and 0.8 micrometers which is right in the middle of the HST band pass. The key to optical detection of a planet in reflected light is an aperture sufficient to place the planet at least 5 Airy rings distant from the star. Only then can coronagraphic and apodization techniques achieve adequate suppression of diffraction, given the expected contrast ratio of 1000 for a solar system twin. At 10 parsecs, a one-AU orbit gives a maximum angular separation of 0.1 arcsec, placing it at the eighth Airy ring.

2.3.2. The Formation and Evolution of Galaxies

The Hubble Deep Field, which provides the deepest optical view of the Universe ever seen, is arguably one of the most important scientific achievements of the twentieth century. Two facts were immediately clear from the HDF. First, HST can resolve galaxy-sized systems at high redshifts. Second, "the Universe at high redshift looks rather different than it does at the current epoch" (Williams et al. 1996). Subsequent papers have shown that Williamsí preliminary conclusion is indeed correct, and, if anything, is an understatement of the strong evolution of galaxy morphology with redshift. At redshifts greater than 1 the Hubble sequence progressively fails to describe galaxies (c.f. Figure 2.3.). As the redshift increases, we observe an increasing population of merging galaxies, peculiar galaxies, and irregular galaxies (Abraham et al. 1996, van den Bergh et al. 1996, Marleau, F.R. & Simard, L. 1999).

Another striking result from the deep HDF images is that the differential galaxy counts in both F450W and F814W flatten at the fainter end. Furthermore, the NICMOS deep images in the HDF did not reveal a new population of faint IR galaxies. Taken together, these facts suggest that we may have already seen into and beyond the era when galaxies first formed.

The HDF was recorded with a camera with 0.1 arcsec pixels. New cameras with higher sensitivity and smaller pixels already planned to be installed will reach the diffraction limit of the existing 2.4-m aperture, doubling the resolution at 0.5 mm wavelength. With the HST10X addition, the same new cameras will obtain a further increase in resolution to 0.012 arcsec. Because of the increased collecting area, it will still be possible to reach the same limits of very faint surface brightness, despite the much higher resolution. The same advance in resolution as was made by the current HDF over previous ground-based images can be made again; the HST10X HDF images will be nearly an order of magnitude sharper than those of the current HDF. The telescope will have the potential to see details in Ly-alpha of the first star forming regions, with redshifts as high as 14 at 1.8 micrometers.

Establishing the era of galaxy formation is one of the primary goals of NASAís origins program. To achieve this goal we need: 1) I, Z, H, and J band images that are deeper than the HDF, and 2) redshifts of faint HDF galaxies. HST10X will provide both the deep images and redshifts that cannot be obtained from the ground. One orbit HST10X images with the ACS will go deeper in the I-band than the HDF, and allow images in the Z band, which cannot be obtained with WFPC2. HST10X H and J band images with the IR channel in WFPC3 will be far deeper than NICMOS images, and will be deeper than H and J band images taken with terrestrial 8 and 10-m telescopes.
 
 

In Figure 2.3. the bright spiral galaxy next to the star in the central strip is at z = 0.517. The next three brightest spirals in the central strip, counterclockwise from the top left, have redshifts of 0.47, 0.319, and 0.32. The two brightest galaxies in the lower strip have redshifts of 0.454 (top left) and 0.432 (bottom right center). The cluster of four faint "lumps" just above the "HST" in the label have a redshift z = 3.216. The linear chain-like galaxy in the top strip that is near the bottom edge and right of center has z = 1.02. The similar, but fainter blue galaxy near the top and left of center has z = 2.8. HST10X would resolve the structure in the fainter galaxies in Figure 2.3., and would be able to measure their redshifts.

HST10X with STIS will be able to measure the spectra of faint HDF galaxies that cannot be measured from the ground because of seeing limitations and because of the bright airglow lines in the NIR.

Finally, as illustrated in Figure 2.2., HST10X will be able to study young evolving galaxies with ~3.7 times the spatial resolution and ~14 times the spatial information of the dithered HDF images (FWHM = 0.14¢¢ in F606W images).

2.3.3. Cepheid Distances and Large Scale Flows within 100 Mpc

The rate of expansion of the Universe and the large scale distribution of dark matter, two outstanding problems in astronomy, are inextricably bound to one another. At one time we thought that the we could measure the local expansion rate of the Universe, i.e. the Hubble constant, H0, simply by dividing the recession velocity of the Virgo cluster by the distance to the Virgo cluster (~17 Mpc). We subsequently learned that galaxies have peculiar motions with respect to a local inertial frame. More importantly, we have shown that the Local Supercluster is far from virialized, and that galaxies and groups of galaxies, including the Local Group, are still falling into Virgo. Most recently, we have established that there are galaxy flows on scales of tens of Mpc, evidently driven by the large scale, inhomogeneous distribution of dark and luminous matter (c.f. Figure 2.4.). Measuring the Hubble constant has proved to be a challenge indeed.

The H0 Key Project has approached this problem by using Cepheids to make primary distance measurements to galaxies that calibrate several secondary standard candles such at the Tully-Fisher relationship, the globular cluster luminosity function, the planetary nebula luminosity function, and the luminosity of the tip of the red giant branch. The primary measurements are made within the Local Supercluster, and then extended to larger distances via the secondary calibrations. The success or failure of this approach hinges on avoiding systematic errors (e.g. Malmquist bias) when using the secondary standard candles. A direct measurement of Cepheid distances out to 100 Mpc would be far preferable.

The HST has proved to be a "Cepheid Machine" par excellence, measuring Cepheid distances out to 23 Mpc (NGC 1425 in the Fornax Cluster). Three factors revolutionized the detection and photometry of Cepheids: HSTís resolution, limiting magnitude, and optimal scheduling. The observing strategy used during the HST Key Project on the Extra-galactic Distance Scale gave an "effective resolution" that was between the 2-pixel resolution of each epoch and the ~1 pixel resolution of all epochs combined. If we suppose that the effective resolution was ~1.4 WFPC2 pixels, the resolution gain with HST10X will be ~4.5. Consequently, HST10X will be able to search galaxies for Cepheids at 100 Mpc with the same spatial resolution in the galaxy that HST presently realizes at 20 Mpc. Cepheids at 100 Mpc will appear 3.4 magnitudes fainter than Cepheids at 23 Mpc. Table 2.4. shows that HST10Xís one-orbit limiting magnitudes in F555W and F814W are respectively 3.65 magnitudes and 4.1 magnitudes fainter than WFPC2 limiting magnitudes. With HST10X, we can use Cepheids for a one-step, direct mapping of large-scale flows to 100 Mpc. These measurements will give H0 directly, averaged over ~100 Mpc. Simultaneously, the measurements will delineate large scale flows and the large-scale distribution of dark matter that drive the flows. The large-scale distribution of dark matter now bears directly on the birth and early evolution of galaxies that will be observed with HST10X. Thus, HST10X observations of nearby large-scale flows and high-redshift galaxies will be key elements of a coherent picture of the origin and evolution of galaxies and the Universe.

2.3.4. Origin of Stellar and Planetary Systems

Two of the most stunning results from HST are the discovery of proplyds (proto-planetary disks) in the Orion Nebula ó young stars surrounded by circumstellar disks embedded within or silhouetted against the bright H II region (e.g., OíDell & Wong 1996; Bally et al. 1998) ó and the stunning "Pillars of Creation" ó isolated ëelephant trunksí of dense molecular gas that have been overrun by the expanding H II region powered by a cluster of hot stars in M16 (cf. Hester et al. 1996). In both cases, the superior resolving power of HST has provided new insights into star formation, the evolution of proto-planetary disks, and the structure of interstellar gas. The HST-WFPC2 images show that it is necessary to obtain at least 50 milliarcsec resolution in order to trace the disk structure and the disk-jet connection in proplyds in Orion. However, a factor of 3 improvement is necessary to locate and study similar objects in the varied environments of other star-forming regions within 1 - 2.5 kpc of the Sun.

The vast majority (> 90%) of stars currently forming within ~2.5 kpc of the Sun form in Orion-type environments. If planetary systems are common in the Galaxy, they must be able to form and survive within these surroundings. Studies must therefore show that: (a) proto-planetary disks are common in OB associations; and (b) planet formation can occur in the face of energetic processes (e.g., photo evaporation, stellar winds, supernovae) that destroy disks. Questions we might address include:

1. How does planet formation depend on the Initial Mass Function (IMF) and massive stellar content?
2. How does disk survival depend on distance from the massive stars?
3. What are the disk properties as a function of position within the star-forming regions?
4. Considerable important and unique information about proto-planetary disks can be gathered from UV/optical/NIR instruments of HST10X.

1) Compiling a census of circumstellar disks in major star-forming regions within 2.5 kpc of the Sun. Optical imaging from space of dark disks surrounding nascent stars against the bright background emission of ionized gas is an efficient method for finding and studying proto-planetary disk systems at sub-Solar System scales (e.g., Bally et al. 1998; Fig. 4). Extending these studies to distant H II regions requires high spatial resolution, and could be combined with IR studies of embedded sources in nearby clouds that have not yet been exposed by the presence of massive stars in the vicinity.

2) Studying the physics of disk/planet formation and survival in star-forming environments. The high angular resolution of space imaging allows us to probe important physical scales in diagnostic tracers of gaseous flows and mass loss. In particular, is the disk-destruction time scale longer than the time scale to form ~1 cm size bodies in circumstellar disks that are resistant to photo evaporation?

3) Resolving the relationship between massive stellar content and the IMF/star formation efficiency. For example, do massive stars trigger or hinder solar-type star formation in surrounding clouds?
 

Figure 2.4. shows protoplanetary disks in Orion are seen against the background emission in this optical HST-WFPC2 image provided by J. Bally. The object near the center is clearly being disrupted by the energetic winds and photon field of nearby massive stars. Such processes are predicted to destroy the disks on relatively short time scales. The object in the upper-left, on the other hand, appears to be a foreground star/disk system which is evidently outside the influence of the destructive forces of the massive stars. High-resolution imaging of such disks against the bright background emission could reveal the presence of gaps in the disks that may indicate planet formation has occurred.

2.3.6. Large Scale Structure

HST10X-COS/STIS will address the UV spectroscopy science goals described in the recent "white paper" issued by the NASA-chartered Ultraviolet-Optical Working Group (UVOWG) (Mike Shull, Chair) titled, The Emergence of the Modern Universe: Tracing the Cosmic Web (Shull et al. 1999; see http://casa.colorado.edu/~uvconf/UVOWG.html)². A brief summary of these science goals is contained here using some text and figures from the UVOWG report. The new science will leverage the exciting results of surveys by MAP (microwave background at 1º scale), the Sloan Digital Sky Survey, and GALEX (discovery of ~10^5-6 QSOs in the magnitude range 18 < m_B< 20).

The current paradigm of galaxy formation invokes gravitational collapse out of a "cosmic web" of dark matter and baryons distributed throughout the intergalactic medium (IGM). The IGM plays the same role for galaxy formation as the interstellar medium plays for star formation. The Sloan Survey will provide redshifts for ~10⁵ QSO targets that will be used as background sources for mapping out the large-scale structure from z = 2 - 0 of the IGM via absorption by the Ly-alpha forest. The improved spectroscopic throughputs in the UV will enable high spatial frequency sampling of QSO sight-lines, with the goal of producing an IGM baryonic survey on sub-degree angular scales, comparable to that of the MAP explorer and to the structure seen in galaxy surveys. In doing so, we will connect the high-redshift seeds of galaxies and clusters with the distributions of galaxies and the IGM in the modern epoch.

Ultraviolet spectra at moderate redshifts that map the Ly-alpha forest on sub-degree scales contain evidence for the epochs of galaxy formation, metal production, re ionization, and rehashing of the baryons left over from the Big Bang. Theories of primordial nucleosynthesis and cosmological structure formation predict a distributed IGM containing a substantial fraction of the hydrogen and helium synthesized in the Big Bang. According to cosmological N-body hydrodynamic models (e.g., Zhang et al. 1997) gas in the high-redshift IGM begins to collapse into the filamentary web of dark-matter potential wells. The first collapsed objects ("proto-galaxies") may form between redshifts z = 10 - 20, and the first galaxies and QSOs are probably present by z = 5 - 10. The universe at redshift z > 5 appears nearly opaque at UV and optical wavelengths, owing to the strong absorption from hydrogen Ly-alpha in the IGM (the Gunn-Peterson trough).

Based on recent galaxy redshift surveys, astronomers have detected the existence of an organized large-scale structure in the galaxy distribution, which takes the form of large filamentary walls and "empty" voids. By 2010, these galaxy surveys will outline the distribution of luminous matter in fine detail, but the dark, gaseous universe (the IGM) will remain largely unexplored at z < 1.65. (At z > 1.65, the Ly-alpha line is red shifted into the visible band, although several key metal transitions at l < 1216 Å remain in the UV.) Theoretical models suggest that studies of the H I and He II Ly-alpha forest of absorbers in QSO spectra should probe the large reservoir of gas left from the major epoch of structure formation. In fact, the intergalactic Ly-alpha absorbers persist down to very low redshifts, and observations from HST show that many Ly-alpha clouds exist in voids as well as in filamentary walls.
 

Figure 2.5. Large-scale Cosmological Structure	Figure 2.6. Large-scale Filamentary Structure

Figure 2.5. Large-scale cosmological structure, consisting of filaments of galaxies surrounding voids, is seen in the CfA2 redshift survey (Huchra 1999). This "pie-diagram" shows the distribution in recession velocity and right ascension of bright galaxies and four Ly-alpha absorbers found by HST-GHRS toward Mrk 501 and Mrk 421 (Penton, Stocke, & Shull 1999). Evidently, the voids are not entirely empty: two Ly-alpha clouds lie in voids, with the nearest bright galaxies more than 4 Mpc away.

Figure 2.6. Large-scale filamentary structures in both hot and warm baryons are predicted in numerical N-body hydrodynamical models (cf., Cen & Ostriker 1999). Mapping these structures in absorption requires moderate-resolution UV spectra toward background quasars at m_B @ 18 - 20.

Studies of the He II Ly-alpha forest are particularly effective at probing the lowest-density regions of the baryon distribution, while the H I Ly-alpha lines at redshifts z < 1.65 (e.g., Bahcall et al. 1996; Jannuzi et al. 1998) may be used to follow the hydrogen structures down to the present epoch. In combination, these two diagnostics allow astronomers to follow the interplay between the formation of galaxy structures and the IGM. They can also be used to study mass exchange ó the depletion of the reservoir of intergalactic gas into galaxies, and the flow of mass from galaxies back to the IGM through galactic winds and tidal stripping.

1. Hubble Space Telescope, Optics Performance Enhancement, Preliminary Feasibility Study Summary Report, http://www.pha.jhu.edu/groups/hst10x

2. UV-Optical Working Group (UVOWG) White Paper, (Mike Shull, Chair) titled, The Emergence of the Modern Universe: Tracing the Cosmic Web (Shull et al. 1999: see  http://casa.colorado.edu/~uvconf/UVOWG.html

Abraham, R. G.., Tanvir, N. R..., Santiago, B. X.., Ellis, R. S.., Glazebrook, K., & van den Bergh, S. 1996, MNRAS, 279, L47
Bahcall, J. N., et al. 1996, ApJ, 457, 19
Bally, J., et al. 1998, AJ, 116, 293
Cen, R., & Ostriker, J. P. 1999, ApJ, in press, astro-ph/9806821
Crampton, D., Cowley, A. P., & Hartwick, F. D. A. 1987, ApJ, 314, 129
Hester, J., et al. 1996, AJ, 111, 2349
Huchra, J. 1999, CfA-2 Redshift Survey, privately released
Jannuzi, B., et al. 1998, ApJS, 118, 1
La Franca, F., & Cristani, S. 1997, AJ, 113, 1517
Marleau, F.R. & Simard, L. 1999, astro-ph/9807223 22 Jul. 1998 (To appear in ApJ.)
OíDell, C. R., & Wong, S. K. 1996, AJ, 111, 846
Penton, S., Stocke, J. T., & Shull, J. M. 1999, AJ, submitted
Shull, J. M., Savage, B. D., Morse, J. A., et al. 1999, astro-ph/99XXXX
van den Bergh, S., Abraham, R. G.., Ellis, R. S.., Tanvir, N. R.., Santiago, B. X., & Glazebrook, K. G.. 1996, AJ, 112, 359
Williams, R. E.., Blacker, B., Dickinson, M., Dixon, W. V. D., Ferguson, H. C., Fruchter, A. S., Giavalisco, M., Gilliland, R. L., Heyer, I., Katsanis, R., Levay, Z., Lucas, R., McElroy, D. G., Petro, L., & Postman, M. 1996, AJ, 112, 1335
Zhang, Y., Meiksin, A., Anninos, P., & Norman, M. 1997, ApJ, 495, 63

.