A SUMMARY OF THE CORE SCIENCE CASE FOR AN ARCHIVAL PURE PARALLEL SCIENCE PROGRAM WITH HST DURING CYCLE 7

1. INTRODUCTION

• Young stars and star forming regions in our Galaxy and other nearby galaxies.
• The stellar content of galaxies (including our own) in the local Universe.
• Large scale structure in the universe and the distribution and evolution of galaxies.

By its very nature, parallel observing with HST presents the astronomer with only two "simple" choices: since the pointing direction, the prime instrument, and the total exposure time for the pointing are givens, one merely has to decide in what mode the other instruments will be used and how the total exposure time will be broken down by filter, grism, etc. Further constraints are imposed by the special requirements for observing with NICMOS. The need for NICMOS exposures to be dithered will have a significant impact on parallel observing when NICMOS is prime but, with the use of the FOM (Field Offset Mirror), can be accommodated when NICMOS is operated in the parallel mode. In addition, while NIC 1 and NIC2 are nearly parfocal and thus can accumulate good data simultaneously, NIC 3 is badly out of focus. When the PAM is adjusted to its maximum travel to bring NIC3 close to a usable focus, NIC1 and NIC2 are useless.

2. THE PARALLEL OBSERVING PROGRAMS

Galactic Program: This program will be for all pointings with a galactic latitude less than 20 degrees. Its primary objective will be a search for low mass stars and sub-stellar objects as well as for UV bright stars. The study of large galactic emission line regions, which normally would be part of this program, have been put into the "special" category because of a very different set of instrument parameters.

Extra-galactic Program: This program will emphasize deep, slitless spectroscopic surveys in the optical and near-IR over a large area (large for HST, that is) and wide redshift range. It will become operative for all HST pointings above b=|20| unless the pointing is at a "special" object. One scientific objective will be to search for and identify emission line galaxies and quasars at high redshift. Simply put, the program will consist of images of different exposure times through a few filters and available grisms on STIS, WFPC, and/or NICMOS.

Special Objects Program: This is the most complex of the programs as the details of the observing parameters depend on the object being observed by the prime instrument. It encompasses Galactic and extra-galactic targets, the latter at both high and low redshift. The following objects to which HST will point are considered in the "special" category: Large Galactic Emission line regions; Galactic globular clusters; the Galactic bulge; local group galaxies, including the Magellanic Clouds; non-local group galaxies that are > 4 arcmin in size (a scale set so that at least one of the parallel instruments will be able to image the galaxy at not too great a central distance when HST is pointing at it); galaxies, clusters, and quasars with redshifts between z~1.4 to 3.6.

3. THE SCIENCE

At low galactic latitudes extinction becomes an important issue for optical instruments while crowding will begin to limit the effectiveness of grism surveys because of overlapping images. The main "generic" program will be a search for low mass stars and sub-stellar mass objects. A second, more specific program will be an imaging survey of large Galactic H II regions with narrow band filters designed specifically for emission line work on WFPC2, NICMOS/NIC3, and STIS. Several parts of the "special objects" program, discussed in §C, are also Galactic science, e.g. globular clusters and the Galactic bulge.

1. The Search for Low Mass Stars

The objectives for STIS and WFPC2 for generic low latitude observing will be similar to that for NICMOS as described below -- search for and identify stars at the low end of the main sequence. With STIS the program will take advantage of its unique ability to do slitless spectroscopy in the ultraviolet and in the far red, i.e. just to the blue of NICMOS grisms. The red grism will be used on all exposures of one orbit or less while the G230LB grism will be added for two orbits and longer exposures. Short, direct images will accompany the grism exposures to aid in identification. For 3 orbits and longer, imaging with 50CCD and LP will be added. The red grism, which reaches to 1µm will be a valuable adjunct to NICMOS in the identification of red stars. WFPC2, with its large complement of filters, will employ a progressively larger number of them as any given parallel opportunity increases in length.

The great sensitivity of NICMOS/NIC3 in a slitless, spectroscopic mode, in spite of the less than optimal focus that is obtainable with NIC 3, combined with the distinctive near-infrared spectra of recently discovered brown dwarf candidates, has led to the formulation of a new, powerful search strategy for low mass objects that can be employed with NICMOS in parallel observing mode. With the recent discovery of GL 229B (30-55 M_Jupiter), the existence of brown dwarfs as a stellar class has become a reality. Nevertheless the transition region linking these objects with the lowest mass stars remains largely unexplored in part because of their intrinsically low luminosities (e.g., GL 229B is only 6.4x10 -6 L). Thus, only the nearest examples could have been detected and only small volumes of sky have been searched. The discovery and characterization of both brown dwarfs and transitional objects (e.g., GD 165B) as would contribute significantly to such topics as star formation, physical and chemical processes in the region between stellar and planetary atmospheres, "dark matter", and galactic structure.

As expected from model atmospheres the increasing presence of CH 4 and H 2 O with decreasing effective temperature causes profound changes in the near-infrared spectrum. In fact, the near-infrared colors of GL 229B are bluish (e.g. J-K ~ - 0.1) while the optical minus infrared colors (e.g., R-K = 10.8) are extremely red because of its low effective temperature (~900K). Particularly distinctive of brown dwarfs is the sharp absorption edge due to methane at 1.6µm. The G141 grism would give the most diagnostic information about cool stars. Therefore, for b between |4| and |20| degrees, a NICMOS parallel survey program with the G141 grism would have the goals of 1) discovering stars at the very low mass end of the main sequence; 2) the closely related issue of discovering brown dwarfs. Not only would the survey result in the discovery of candidate low-mass objects, but it would also indicate the spectra type of any candidate. This program can be executed with only one orbit parallels with emphasis on doing as many fields as possible rather than going very deep. A direct image using F160W would be included to register positions for the grisms extractions. This program will be complementary to the high galactic latitude parallels with the G141 grism as the high latitude program may also result in the discovery of cool stars while this program may also discover QSOs and other extra-galactic emission line source. This survey would not be particularly effective at latitudes less than |4| degrees because the high degree of crowding would make interpretation of the NICMOS data very difficult.

2. Emission Line and Star Forming Regions and Molecular Clouds

Emission line imaging of well resolved high surface brightness nebulae with WFPC2 on HST has been very successful both in terms of public appreciation and scientific return. The color mosaics of images of the core of M42, of the elephant trunks in M16 and of a wide variety of Galactic nebulae from HH objects to SNRs have brought valuable data on the small scale structure of the ISM and on the earliest stages in the life of protostellar objects (proplyds). Parallel pointings of greater than 1 orbit within large Galactic HII regions, SNRs or planetary nebulae will allow imaging in several emission line filters and give us the ability to study spatial variation of ionization conditions, the role of shocks and scattering by dust, and to search for the presence of small scale structures (e.g. proplyds and neutral globules). Imaging in a number of lines, including weak diagnostic ones (such as [O III] 4363Å), will be the equivalent of doing "imaging spectroscopy" at 0.1" resolution.

The narrow band filter complement of WFPC2 is well suited for emission line imaging. The data can be accurately flux calibrated. A proposed priority listing of filters with indicative exposure time ratios to achieve a similar signal-to-noise is given in Appendix A. This choice of filters will permit study of both high and low ionization regions, extinction variations, the scattered light component, and yield estimates for temperature and abundance variations. Ground based studies have in fact shown that line ratio maps from strong lines can be used to derive fairly reliable (±0.2 dex.) abundances.

STIS' [O II] and [O III] filters with pixels of similar size to the PC will also provide valuable deeper emission line imaging over a restricted field.

NIC3 has several filters which bring complementary data on the gaseous emission more applicable to heavily embedded regions. For example, the Paschen , H 2 , [Fe II], and He I filters can be used to explore embedded gas, shock regions (H 2 and [Fe II]) high density regions ([Fe II]), the higher ionization zones and possible He enriched regions (from He/H). Nearby continuum filters will be needed at the longer wavelengths to effectively subtract the background (e.g. F215N). These observations will also be able to explore gas outflows and young stellar objects.

B. The Main Extra-Galactic Parallel Programs

1. The NICMOS Search for High Redshift AGNs and QSOs

The PWG proposes that the NICMOS Extra-galactic parallel program obtain NIC 3 direct images with the F110W (J) and F160W (H) filters for long (>= 6) pointings, grism G096 and G141 images for intermediate pointings (augmented by short direct images), and additional short direct F160W images for a large number of STIS and WFPC prime targets with single orbit exposures away from the Galactic plane. An evaluation of the relative numbers of grism and deep direct images taken by mid 1998 may lead to an inclusion of direct imaging with at least some of the 5 orbit parallel pointings. This program again underscores the unique value of imaging with NIC 3 in spite of the focus problems. With this data set a number of important extra-galactic problems can be addressed:

a) Active Galactic Nuclei and Quasars

With the large spectral coverage of the combined G096 and G141 grisms and their low background, emission lines associated with AGNs and QSOs can be detected to flux levels as deep as the most sensitive ground-based broadband optical surveys and over a very large redshift range -- from z >> 1 to z >> 9 -- over most of which at least two and usually three strong emission lines can be seen for redshift identification. In particular: (a) Very high redshift (z >~ 5.6) AGNs and QSOs will have their Lyman emission shifted into the range of these grisms, and (b) Since these objects will be found from searches in the near IR, they will be significantly less affected by either intrinsic or intervening extinction then would objects at the same redshift observed at optical wavelengths. This data set will thus allow the investigation of the luminosity function of very high redshift AGNs and QSOs and an assessment of the effects of dust extinction on their luminosity functions. The strategy will be that each 2-3 orbit pointing will use only one of the grisms, alternating between the two, while each 4-5 orbit pointing will be divided equally between the two grisms.

b) Galaxy Evolution

The combination of the F110W and F160W imaging together with the grism images will provide a valuable database for the investigation of several important aspects of galaxy evolution: a) History of Star formation: One or both of the star-formation rate indicating emission lines H-alpha and [OII] can be seen over the entire redshift range from 0.2 to 4.1, which, augmented by the J-H images accompanying the grism images, will provide a valuable database for mapping in more detail than has been possible the star formation history of galaxies. b) The deeper direct images will augment existing optical imaging surveys (e.g. the Medium Deep Survey) by providing color and morphological information which samples redder rest wavelengths than the optical surveys and hence is more sensitive to the older stellar populations at redshifts where the optical data is sensitive primarily to the star formation activity at that redshift. The J-H color is particularly interesting for galaxies in the 1.5 < z < 3 regime where the H & K break or the Balmer discontinuity is being shifted through these two passbands. The large solid angle (compared to a single NIC3 frame) will also supplement the number/morphological database obtained from the GTO NICMOS 3 HDF campaign by providing statistical information on the brighter, rarer objects at resolutions which, even at the slightly degraded spatial resolution of the NICMOS 3 frames, will be significantly better than ground-based data.

2. WFPC2 Extra-Galactic Parallel Program

a) A Broad Band Survey

We suggest that WFPC2 parallel observations be obtained in random high-latitude fields with the objective of studying the properties of faint galaxies, specifically their size, color and morphology in different bands. Such a program would build upon the success of WFPC2 pointed and parallel observations, using the HDF filter set for continuity and maximum depth. The V and I filters also overlap the bulk of the Medium Deep Survey. This program would be able to map the number, angular size, color, and morphology of galaxies at cosmologically significant redshifts. It will complement the deep STIS parallel programs (described below) for high latitude fields.

While deep observations with WFPC2 have already allowed a program similar to this one to extend to very faint magnitudes, there is less information on relatively bright (V ~ 24 mag) objects, which number only a few tens per field, in multiple colors (UBVI). The U and B filters will add key information on dust content, stellar populations, and recent star formation for the galaxies imaged. Some specific goals of the program we are proposing are:

• Determine number counts vs. morphological type and color, and magnitude-size-color relations in different filters.
• Study the morphology and color gradients of field galaxies to constrain stellar population and dust distribution
• Morphology of intermediate redshift galaxies (z=0.2-0.8) in the UV. This will provide a direct counterpart for the V and I observations of high-z galaxies (rest frame UV), and lead to a better understanding of the evolution of morphology and star formation properties with redshift.
• Study morphology, size and color of the numerous, possibly sub-galactic blue compact objects found at B ~ 24
• Identification of Lyman dropouts. At the magnitudes that can be reached in a few orbits, such galaxies will be very luminous high-redshift objects, and therefore rare (one every few fields) and difficult to find via pointed observations.

b) A WFPC2 Survey in Medium­band Filters F410M and F467M.

The PWG proposes that a high latitude medium-band filter survey be carried out with WFPC2 but with a lower priority than the broad band survey. Its implementation will depend on the progress made on the latter parallel program. This survey with the medium-band filters F410M and/or F467M will use a modest (~10) WFPC2 parallels of 6-8 orbits each plus short exposures in the continuum filters B450 and I814. These filters are unique with / ~ 3.5% and central wavelengths similar to some of the Strömgren filters. Over 600 Cycle 4-7 shorter exposures were taken in these filters for a variety of purposes that cannot achieved by other means, including imaging of [O II] emission in radio galaxies and Abell clusters at z~0.1, and imaging of faint compact Lyman emitting objects in two known clusters at z~2.4, as well as parallel searches in three random fields for such objects.

The following principle scientific issues could be addressed with this program:

1) These filters can find faint and compact O­II] emitters at z~0.1 or ~0.25, respectively. A search for faint blue compact dwarf galaxies at intermediate redshifts is important to address the steepness of their LF, and see how and why the population of dwarf galaxies recently faded with time. This a search can be done at random, or when known nearby objects (galaxy clusters or radio galaxies) at z~0.1 or ~0.25 are the primary target.

2) A search for faint compact Lyman emitting objects in groups and clusters at z~2.36±0.06 (F410M) and z~2.85±0.07 (F467M). The width of these medium band filters is essential to slice up the redshift distribution of faint star-forming objects at z~2.4-2.9. Their N(z) recently showed significant structure, possibly with a similar redshift separation as the "spikes" seen in more nearby redshift surveys (at z/z ~ 0.04). Hence, narrow band filters are not efficient for such Lyman searches, since they would largely look in between any such structures. About ~150 faint compact Lyman emitting objects are expected at z~2.4-2.9, allowing one to: (a) derive the LF of faint compact Lyman emitters, which may be quite steep, and its evolution with time when compared to that at z=0; (b) determine their Lyman fluxes, SFR, AGN fraction, and the fraction obscured by dust; (c) determine their size distribution and see if these objects have the sizes of today's galaxies, or are significantly smaller.

3. STIS Extra-Galactic Parallel Programs

a) High Latitude STIS Slitless Spectroscopy

One of the goals of observational studies of galaxy evolution is to determine the star formation rate as a function of lookback time and environment. Such a determination of the star-formation history of the universe will place strong constraints on models for the formation and early evolution of galaxies. Ground-based imaging and redshift surveys, in combination with the HST Medium Deep Survey key project, and the Hubble Deep Field, have begun to build up a consistent picture of galaxy evolution. In this scenario, the population of early­type galaxies has undergone very little star-formation at redshift z<1, while a population of morphologically complex and active star-forming galaxies was numerous at z>0.4, but have disappeared by the present epoch. Although a population of star-forming galaxies has recently been discovered at z>3, the relatively low number of UV-dropout galaxies in the HDF indicates that the bulk of star formation takes place at z<3. Although there is still considerable debate about this picture, it is clear that the poorly-studied redshift range 1<z<3 is of critical importance.

In addition, recent theoretical work has managed to reproduce the small scatter in the color of cluster elliptical galaxies with a hierarchical merging model. In such a picture, the redshift range of greatest interest might also be 1<z<3 rather than higher redshifts. By providing unprecedented sensitivity in the wavelength range 7000<l<10000Å, slitless spectroscopy of random fields with the STIS camera in the G750L mode could detect [O II] emission caused by star-formation at z<1.7. These data would also be sensitive to Ha emission at z<0.5, [O III] emission at z<1.0 Lyman emission at 3.5<z<7.0. We propose that such a program be carried out for ~85% of the two orbit or longer visits when instruments other than STIS are prime. As a special case, when the prime observations are of objects known to be in the redshift range 1.5<z<3.5, we suggest using the STIS in the G430L mode to attempt to detect Lyman at these redshifts in their neighborhood.

b) Measuring the Cosmic Shear with STIS

"Cosmic shear" is the measurable distortion of images of high-redshift galaxies caused by the tidal gravitational field of the large-scale matter distribution in the Universe. As the light bundles from these high-redshift galaxies propagate to us, they are affected by the integrated tidal fields along their respective lines-of-sight. Since the intrinsic orientations of galaxies are random, any net alignment (or net ellipticity) beyond Poisson noise can be attributed to the cosmic shear. A measurement of the cosmic shear would directly yield a direct measure of the power spectrum of cosmic density fluctuations, without any reference to the relation between dark matter and luminous tracers. In this respect, cosmic shear as a cosmological tool is comparable only to the measurement of the cosmic microwave background. The rms shear in a given aperture can be calculated directly in terms of the power spectrum of the density fluctuations, which in turn depends on the cosmological parameters and the matter model.

Single orbit parallel imaging opportunities with the STIS unfiltered CCD at b >|20| can yield a statistically significant detection and quantitative measurement of the amplitude of the Cosmic Shear. Its sensitivity and unique image quality, make the instrument ideally suited to measuring a shear on a 1 arcmin scale. The expected rms of the cosmic shear on this angular scale is about 3%, in which case a highly significant detection will be possible with ~20 single orbit frames. This detection alone will provide a spectacular scientific success. A precise (say, to 10% accuracy) measurement of the rms will require a few hundred frames and will permit the determination of the normalization of the power spectrum of the cosmic density fluctuations and the determination of higher order moments of the shear distribution. Determination of these moments will allow the study of the non-Gaussian nature of the large-scale matter distribution due to non-linear gravitational clustering. These single orbit images will be part of other extra-galactic parallel science programs that require multi-orbit images of the same field.

c) A Deep High Resolution Parallel Survey with STIS

As noted above, we have recommended that 85% of the two orbit or longer high latitude visits, parallel observations with STIS be devoted to a slitless spectroscopic survey of the Universe. We propose to utilize roughly 15% of the parallel opportunities which are longer than four orbits for a high latitude imaging survey with the STIS CCD. This will provide imaging of the faintest possible stars and galaxies with the best possible HST spatial sampling in the optical. This should optimize the ability to achieve star-galaxy separation and galaxy classification at the faint end of the apparent luminosity distribution. This STIS parallel survey would be with the clear 50CCD (a pseudo V) and with the 50CCD+LP filter (a pseudo I). Thus, although only a fraction of the clear field is imaged through the LP filter, these additional data will add some color information to the exceptionally deep images this program will obtain with the 50CCD alone. In the 10 to 12 pointing opportunities predicted to obtain during the course of this program, 0.002 ¤ ° will be imaged down to an equivalent V of 27-28 mag with ~0.1" FWHM images. An estimated 2000 faint field galaxies and at least several dozen galactic stars will be detected.

The broad science goals for this program are as follows with approximate V and I magnitudes given:

• Study the faint galaxy population, their colors and counts as a function of morphological type for V~<28 and I~<28 mag, thereby constrain galaxy evolution beyond the levels where it can be addressed spectroscopically.

• Study the sub-kpc rest frame morphology and color gradients of faint field galaxies to constrain stellar populations and dust. Study the compact faint blue objects for an equivalent B<28 mag (whose expected median scale-length is ~0.1-0.3"), which requires the full STIS sampling.

• Star counts with reliable star-galaxy separation as a function of Galactic latitude down to V~29-30 and I~29 to constrain Galactic halo/bulge/disk models. Stacks of 6 to 8 orbits will also give us the opportunity to find faint stars that are variable on time scales of ~6 hrs down to V=28-29 mag.

• If there are any 6-8 orbit STIS images in 50CCD at low ecliptic latitudes then splitting them up in two stacks of 3-4 successive orbits each should give good enough CR-rejection that faint solar system objects can be detected through their proper motion.

This part of the core parallel program we are proposing would utilize the diffraction limited imaging capabilities of HST in the optical and near-IR (NIC 2 will be used for this program because of crowding and need to go deep for stellar images) and the very high throughput and overall QE of the STIS CCD (esp. compared with WFPC2) to study the stellar content of our own and other galaxies in the local group. This special class of objects will include both the Galactic bulge and a subset of Galactic globular clusters. Also included will be a short list of large (>4 arcmin) galaxies more distant than the local group. Because of the spacing of HST's instruments in the focal plane this program would be activated whenever HST's prime pointing with a given instrument is within a predetermined distance from the center of a pre-specified list of galaxies, star clusters, or the Galactic bulge. Otherwise, either the extra-galactic or galactic programs will become operative. An important overall goal of this part of the program is to determine star formation rates as a function of the mass, metallicity, and age of the parent system. Results of this program would bear strongly on the interpretation of the light of distant galaxies and on understanding details of the process of galaxy formation.

For galaxies and star clusters with (m-M) < 23, parallel imaging will be able to determine colors, magnitudes, and luminosity functions (hence mass functions) for giants and main sequence stars in a wide variety of environments. In a few orbits, for example, in the near-IR with NIC 2 it will be possible to reach the hydrogen burning limit in some star clusters and Galactic bulge. In contrast to the situation for globular clusters, in dwarf spheroidal and irregular galaxies, such as the SMC and Ursa Minor, the initial mass function is expected to be unaffected by dynamical processes because of these latter systems' low stellar density. We should be able to determine the IMF over a range of 2 dex in metallicity, and over 3 dex in galaxy mass.

For objects with (m-M) > 23, i.e., primarily M31, its companions, and beyond, this core parallel program emphasizes the evolved phases of stellar evolution in old populations, and the young main sequence (when present). The goal is to constrain the star formation history of stellar halos, and to explore the diversity of old populations. In the cases where the target has active star forming regions, the imaging would also probe the high mass end of the mass function and the spatial distribution of star forming regions.

In old (>10 Gyr) populations, the metallicity distribution of stars in a galactic stellar halo can be determined from the spread in the giants' colors. This can be compared to chemical evolution models, and may also constrain the fraction of mass acquired via mergers. The tip of the giant branch, a valuable distance indicator in the red and near-IR, and AGB stars in general can be studied to (m-M) = 30. Luminous AGB stars would be strong evidence for an intermediate age stellar population. NICMOS imaging can give bolometric magnitudes from the H magnitude alone, and can confirm whether carbon stars (a tracer of intermediate age populations) are present.

The use of the grism mode in STIS (G750L) can give classification­quality spectra of the evolved late-type giants, permitting separation of carbon and M stars for many local group galaxies. This will be an important adjunct to the work that has been done from the ground on the Magellanic Clouds and a few of the nearest dwarfs spheroidals. One orbit grism observations of local group and special large galaxies will be carried out for prime (non-STIS) observations of 2 orbits or more towards these objects.

The study of their stellar content of local group and other large galaxies as well as Milky Way globulars with WFPC2 via multi-band imaging can be subjected to more fine tuning than is possible with STIS or NICMOS because of the large number of filters available. This variety of filters permits a detailed study of the young stellar content of these systems in addition to their older stars. Single orbit V images can normally go deeper than V=27 with a photometric accuracy of ±0.1 mags. However, in medium to high surface brightness areas of galaxies, V images become confusion limited for V>25. Examples of filters and exposure times with WFPC2 are given in Appendix A.

Scientific objectives and filter details for the three categories of objects are given below. These objectives both complement and overlap the objectives of the STIS and NICMOS observations.

For the study of the stellar content of the Magellanic Clouds as well as the closer (. By fitting model atmospheres to the observations, it will be possible to determine T eff , and L and, in many cases, individual reddening values for many individual stars, especially - and this is a crucial point - in star forming regions. These images should reach main sequence stars of 0.9 M in the Magellanic Clouds, and about 1.6 M in M31. It will also be possible to directly detect pre-main-sequence stars, e.g. those with H alpha excess emission, T Taurus stars in the Magellanic Clouds, and Herbig Be/Ae in M31. The result will be a complete characterization of the stellar population, including the main sequence, for stellar ages between 1 Myr and 8 Gyr in the Magellanic Clouds, and between 1 Myr and 3 Gyr in M31.

The third group of special objects will be the predetermined selection of large galaxies outside of the local group that can simultaneously be observed by both the prime instrument and the parallel one. The details of this program will be similar to that for Galactic globular clusters, i.e. observations with BVI filters only and concentration on the old stellar populations. One of its objectives will be the characterization of the globular cluster populations of these large galaxies in terms of their spatial distribution, their color (to get their metallicity) distribution, and their luminosity distribution. This information will yield clues as to the formation and early history of these galaxies and their cluster systems. Furthermore, for galaxies out to ~20 Mpc it will be possible to measure the top of the RGB and AGB giant branches.

---

APPENDIX A

Local Group Galaxies and Star Clusters For NICMOS, the proposal employs the F110W and F160W filters for integrations < 1 orbit, adding the F220W and F237M filters (useful in the study of late-type stars) for longer integrations. The STIS program takes advantage of a clear aperture and the long pass (LP) filter. In cases where WFPC can run in parallel, imaging would use the F555W (V) and F814W (I) filters.

WFPC2 OBSERVATIONAL PLAN

Case 1: High latitude (|b| > 20): Filter priority: U V B I (but V for 1 orbit cases, as U alone is not useful)

Orbits         F606W       F300W      F450W     F814W
   1          2x1000
   2           2x500      2x1500
   3           2x500      2x1500     2x1000
   4           2x500      2x1500     2x1500     2x500
   5           2x500      2x2000     2x1500     2x500

and so on - increase F606W, F814W up to 1 full orbit each, give the rest to
F300W, F450W.  Note that CR splits are still required.

Case 2: Low latitude (|b| < 20): Filter priority: V B I U

Orbits         F606W       F450W      F814W     F300W
   1          2x1000
   2           2x500      2x1500
   3           2x500      2x1500     2x1000
   4           2x500      2x1000     2x1000    2x1500
   5           2x1000     2x1000     2x1000    2x2000

and so on - increase all filters, keeping F300W longer than the others.

Case 3a: Milky Way Globular clusters (old population only); Filters B,V,I; exposures TBD

Case 3b: Magellanic Clouds (mixture of populations or, even, mostly young population).

Filter priority: V B U I H a - need different exposure times for expanded magnitude range (young population)

Orbits  F606W   F450W   F300W   F814W   F656N
1st    40,200  60,300  60,300  100,500
2nd        10      10      20   2x1000
3rd       ---     ---    1000      ---   1500
4th+   repeat 1-3 cyclically

H filter needed to reveal stars with strong H excesses (8Å , or 0.3 mag relative to continuum). Stars later than B type with Ha equivalent widths greater than 8 Å are all pre-main-sequence stars. The continuum flux at 6563 Å is interpolated from F606W and F814W (internal accuracy better than 0.05 mag).

Case 4: Nearby galaxies (Outside local group, up to Virgo - a subset of objects > 4')
Goal: study globular cluster systems (LF) and very bright individual stars. Need three filters to separate between single stars and mini-clusters. Essentially BVI with comparable exposure times.

1 orbit:       F606W (CR-split)
2+ orbits:     F606W, F814W, F450W, equal length, each CR-split

Case 5: Galactic HII Regions (exposure times to be used on bright, large Galactic HII regions, SNRs, and planetary nebulae:)

H-alpha   F656N    1.0
[O III]   F510N    2.0
[N II]    F658N    3.0
[S II]    F673N    6.0
H-beta    F487N    3.0
[O III]   F437N   30.0
[S III]   F953N    8.0
[O II]    F375N    9.0
Narrow V  F547M    0.5
Narrow B  F467M    0.5
[O I]     F631N   20.0

A justification for this ordering is that H- explores the bulk gas emission, whilst the high ionization is detected though [O III] and the low ionization in [N II]. The [S II] is a good diagnostic for shocked gas. H- in association with H- allows the extinction to be mapped. The [O II]5007/4363Å ratio is temperature sensitive, whilst [S III] allows more embedded high ionization regions to be studied. [O II] together with [O III] explores abundance variations. Filters chosen to be off strong emission lines can be used to study the scattered light since the nebular continuum can be corrected in these filters using the H- images. [O I] is good for studying (low velocity) shocked gas or cooler regions / PDR's.

---

Copyright Notice

archive@stsci.edu