Dear Readers,
Welcome to issue 6 (November 16, 1998) of the MIPS/IRAC GTO newsletter. The highlight of the past two weeks was the SIRTF SWG workshop in Pasadena on November 11-13. In addition, MIPS had a team meeting on November 10 to discuss their GTO science plans. George Rieke presents highlights from these meetings below. In addition, Bob Gehrz presents an outline for an upcoming white paper and solicits input from interested parties.
The deadline for submissions to the next issue of the newsletter is the 27th of November. It is the day after Thanksgiving, and it promises to be a good day to market your GTO science plans.
Doug Kelly, editor (dkelly@as.arizona.edu)
The scientific objectives of this project are to map the distributions of the stellar populations that are important in galactic chemical evolution and to assay their global chemical contributions to the galactic ISM gas and dust.
A number of stellar populations can be recognized in nearby galaxies based on their spectroscopic and photometric signatures. In nearby spirals such as M31 and M33, these include:
What goes into the ISM from these populations? There are a number of ejecta that can be evaluated with SIRTF:
What is the incentive for doing this work with SIRTF rather than doing it from the ground?
There are several components to the observational strategy. IRAC images will be taken on timescales of days, weeks, months, and years to identify luminous, dusty stars and variable sources and also to define galactic structure (all bands required). MIPS images will be used to locate HII regions and ISM clouds (all bands required). MIPS and IRAC follow-up will be used to determine SEDs. IRS follow-up will be used to assay dust and gas contributions from the different stellar classes. Blink comparison of large image databases will be required in near real-time to determine the follow-up targets.
The following is a possible outline for a one-year study of M31:
Options for the following 1.5 years:
(editor's note: The tables and figures from Dr. Gehrz's presentation
will be presented along with his full white paper in an upcoming issue
of the newsletter.)
To Top
There was a lot of discussion about SIRTF science last week at our team meeting, at the Science Working Group, and at two workshops that followed the SWG. Many of the ideas are stirred up together, so I will just write a summary of interesting points from all three events. I will do so from the point of view of unifying the possible MIPS science programs into just two - Galaxy and AGN Evolution and Young Stars and Planetary Disks. I find it interesting to think of things pulled together in this way because it helps emphasize where there are potential gaps in the overall program, as well as bringing out any possible uses of similar data to gain more than one goal. So here goes!
It became clear during the discussion leading up to the team meeting that a major investment of GTO time in deep surveys would support programs on both galaxy and AGN evolution. The discussion at the team meeting was based on the proposal generated from the perspective of AGN evolution studies that we should survey roughly 2 square degrees, distributed roughly over four regions of sky, and with deep AXAF exposures for as many of them as possible. This approach was felt to give about the right area for galaxy evolution also, and distributing over four fields provides a test of whether any of the areas is "abnormal" compared with the other three. If we use net integrations of about 2000 seconds per point in these fields, we will come close to the confusion limit at 70 microns and will be strongly confusion limited at 160 microns. Originally, IRAC was planning to invest most of their GTO time in a single deep field considerably smaller than the proposed MIPS surveys, but in further discussion at the cosmology workshop in Friday the 13th, they began to think seriously about a 2 square degree survey also. In addition, it was proposed that we obtain something like two fields (50 square arcmin total) very deep at 24 microns, and that we consider covering both of these fields in the 70 micron superresolution mode to see if we can penetrate deeper into the confusion with better sampled Airy disks. IRAC would be interested in such ultradeep fields. Thus, it is possible that we can coordinate closely enough to get thorough coverage in all seven SIRTF bands for both a deep and an ultradeep survey.
The choice of fields is reasonably clear in the northern hemisphere - the HDF, and the Lockman hole field to be observed by AXAF. Other possibilities are the Groth strip, which Ned Wright pointed out is oriented in a way that a single MIPS scan could neatly cover all of it, and WIRE fields if they do not coincide with the above (but it is believed that they do). In the south, the choice is more complex. The AXAF survey field is yet to be designated - they cannot do the HDF south because of a nearby bright star. The HDF south itself is not as useful as the HDF north because it was observed all in parallel mode - that is, where the CCDs pointed there is no NICMOS data, and vice versa.
The most important task to advance the definition of this program is to identify candidate fields as soon as possible (based on ancillary information available and on ultra-low infrared cirrus). We can then try to finalize agreements with IRAC to coordinate on these fields and can start obtaining additional information on them.
Potential additional data would include:
SCUBA
VLA
Deep groundbased optical and near infrared imaging
Xray, if we pick a field without it
21cm as a way to map cirrus
There is a lot that can be done with SIRTF on starbursts. One of the most pressing programs, though, is to study the dependence of starburst properties on metallicity, since an understanding of this dependence is critical to interpret the deep cosmological surveys. One could observe these galaxies with all three instruments, but it appears that MIPS and IRS have the most to do. With MIPS, one would trace out the far infrared emission in the lowest metallicity galaxies, and possible also map low metallicity galaxies to see if the far infrared properties varied locally over the galaxy. With IRS, one would look for mineralogical signatures of interstellar dust and would also characterize the radiaiton field using the fine structure lines.
There are only a handful of galaxies with really low metallicity known, so this program can be well focused and comprehensive. We need to be sure that optical and near infrared spectra are available for them. Coordination with other SIRTF teams is TBD.
The process of enrichment of the interstellar medium can be studied in another manner with SIRTF - to observe the sources in our own galaxy where it is actually happening (or to extend such studies to other galaxies, see white paper prepared by Doug Kelly from Bob Gehrz's viewgraphs, this newsletter). Staying with the more local case, Joe Hora (IRAC), Bill Latter, et al. have thought about a program combining data from all three SIRTF instruments to study ejecta around planetary nebulae and supernovae. Bill showed us some spectacular images of planetaries that emphasized very large structures representing material lost by the star well before it became a white dwarf and formed the planetary. It should be possible to study this material by mapping with IRAC, which has PAH features in some bands but not others, to locate super-heated grains. IRS can be used for spectral maps, which would identify emission lines and tell us about the excitation in the outer structures. MIPS could identify the coldest components of dust in the ejecta and the SED mode could be used along with low resolution IRS spectra to explore the composition of the dust.
Similar studies could be conducted around supernova remnants. In fact, some of the poster talks at the ISO conference in October reported detection of cold dust forming in the outer regions of supernovae remnants, where the expanding metal-rich material was encountering the quiescent interstellar medium.
Assuming that, through a combination of IRAS, ISO, and deep SIRTF surveys we can constrain the evolution of field galaxies in the infrared (and hence in total rate of star formation), it is generally believed that galaxies in rich clusters should have a significantly different development. Processes such as galaxy cannibalism and harassment (the latest term for near encounters that do not result in merging) could make galaxies grow, or form stars more quickly out of their gas, while stripping by the intracluster gas might force them to become quiescent sooner than typical for field galaxies.
These issues can be studied by a series of MIPS deep images of nearby clusters as well as optically selected ones (e.g., without a dominant quasar or super strong AGN) at high redshift. By high signal to noise scan maps, we should be able to locate the infrared sources precisely at 24 microns, then use this information to deconvolve the longer wavelength information. This technique should allow us to detect many individual galaxies even in the crowded environment of a rich cluster. At high redshift, we may have to fall back on the integrated far infrared light of the clusters as the best measurement (hence the emphasis on systems without a central strong AGN).
It is proposed, but not really convincingly confirmed, that rich clusters may have heated intracluster gas contributing to their far infrared emission at a level from a few percent to much larger. A previous white paper discussed the Stickel et al. paper on the Coma cluster that claims to detect such emission, but the results are pretty insecure. A complement to studying cluster evolution would be that the same program would give information about such dust, which if it is present is presumably heated by the high temperature x-ray emitting gas. The prediction is that a residual extended flux should remain after the galaxies in the cluster have been accounted for as described in the preceding paragraph. It is predicted that the lifetime of dust in the x-ray emitting gas should be no more than about 100 million years, so detection of it would indicate a gas/dust source such as recent stripping from galaxies. If such stripping is still occurring, it suggests that gas-rich galaxies are still finding their way into the center of the cluster.
Gas and dust can also be drawn into a cluster by a cooling flow, and there is evidence that cooling flow clusters may extinct their backgrounds more strongly than other clusters. It would be interesting to compare far infrared luminosities and flux distributions to test this hypothesis. Also, we would want to probe clusters of differing x-ray properties to see if the far infrared behavior changed in correlation.
The centerpiece of a study of AGN evolution would be the deep survey of x-ray studied fields described in 1.1. However, to accompany this work we need to pursue programs exploring the relation of various types of activity to each other in the local universe. A number of possible programs were described in the quasar SED white paper. Charles Lawrence proposed a program similar to one of these - to probe the far infrared properties of radio loud quasars and FRII radio galaxies to see if the proposed unification is correct. Right now, this part of the program is, however, perhaps on the weakest ground because, having shown that the radio galaxies are fainter in the far infrared than the matched quasars, the people who have done this with ISO have proclaimed that, of course, it just proves that the accretion disks are optically thick to beyond 200 microns. This isn't to say that there isn't an important program in this general area, just that further work is required to sharpen its definition. Some ideas in this direction can be found in the white paper. Charles looks like a natural collaborator.
Large-scale structures in the cold interstellar medium in early type galaxies may indicate gas that is helping feed both star formation and their central black holes. A spectacular example is the glowing mid infrared barred disk found by ISO in Centaurus A, but other systems are seen in absorption by HST imaging. These "disks" really aren't that simple in structure, because they must assume a configuration that makes them stable in the gravitational field of the host galaxy. Studying them is one way to establish a connection between the host galaxy and central AGN.
IRAC and MIPS have agreed to collaborate on mapping star clusters of various ages from extremely young to fairly young to identify protoplanetary disks and study their dissipation rates. However, at this point the details of the proposed programs differ slightly - probably a positive thing, since the programs appear to be complementary.
IRAC is planning to select a small number of clusters (3 to no more than 6) that have been studied very thoroughly - in fact, they have an ongoing program to improve our understanding of cluster ages, and the outputs of this program would allow them to select clusters that define a very well determined age sequence. Erick Young presented the program he and Charlie Lada have been defining, which would involve studying a larger number of clusters (15 - 20), with an emphasis on learning about the effects of environment (e.g., cluster density) as well as age on the protoplanetary disks. It appears that the work described by John Stauffer on determining cluster ages (independent of SIRTF) would be very helpful in dating not only the IRAC-selected clusters, but also calibrating the dates for the larger MIPS sample. The MIPS sample would give a broader perspective on the evolution of disk structure than could be obtained with a small number of examples.
A second use for the mapping data is to search for very low mass members of the clusters. It becomes impossible to identify such objects in the near infrared because, as one images deeper and deeper, the background stars eventually shine through the enshrouding dust in sufficient numbers that they overwhelm the faint cluster membership. Therefore, all near infrared cluster studies become limited in source identification by the time they reach only moderately far into the brown dwarf regime. However, if sources are identified on the basis of mid infrared excess, we can discriminate against virtually all background stars and isolate likely cluster members regardless of how faint they are. Of course, an issue is whether the faint objects that still have circumstellar material (protoplanetary disks??) have other biases relative to a general sample of the cluster. Still, finding these objects is one of the most promising way to extend knowledge of the initial mass function down into the few-Jupiter range of masses.
Prior work should include both deep imaging and spectroscopy of the candidate clusters so we know as much as possible about each source.
An outcome of surveys of young clusters with MIPS would be a candidate list for very dense, cold cloud cores that could collapse into stars. In addition, Jocelyn Keene proposed a snapshot survey of isolated globules, both at 70 and 24um. The survey could identify the overall pattern of far infrared emission (70um) and search for a more condensed, warm core to a globule (24um). Interesting objects in both the isolated and cluster sample could be studied in more depth with the SED mode, and from the ground in mm-wave spectral lines and submm continuum. Such studies could measure the temperature and density structure of the dark clouds, and hopefully determine how these conditions related to their probably of forming stars.
The proposed debris disk program breaks into a number of subprograms.
First, is to explore with the full power of SIRTF a small number of very prominent systems - beta Pic, Vega, Eps Eri, Fomalhaut, HR 4796A, etc., popularly dubbed "The Dirty Dozen." The first four listed already have been resolved by SCUBA, implying that MIPS images would be of significant interest. The ISO data show that spectra would also be interesting, both to determine mineralogy, to search for water ice, and to provide the detailed spectral energy distribution of the continuum emission. These objects become the templates against which we (and others) would measure our understanding of other debris systems. Jeff van Cleve is a potential collaborator from IRS on this program.
Mike Jura presented a plot of the location of A stars relative to the main sequence (using Hipparchos data to derive absolute magnitudes). The results show a lot of scatter, with the youngest A stars falling systematically toward the zero age main sequence. Although this kind of age dating is not very precise, it may help us understand the trends in an A-star sample of debris disk observations, should we decide to pursue a program in this direction.
Mike Werner and Chas Beichman also suggested we concentrate efforts on a sample of "nearby" stars, within ~ 15 pc, since these are the cases where we have the best chance of detecting and studying very low luminosity disks.
A general challenge to debris disk and protoplanetary disk programs is to make a seamless connection - we should try to understand disk evolution from the youngest clusters into the oldest field stars, rather than approaching the problem in a disjointed way. In addition, we need to narrow down a sample of sources. In particular, to search for low luminosity systems we need to be very careful about avoiding infrared cirrus, which both can provide a background confusion noise to the star being studied, and can be heated by the star to form a "mock disk" that is not a physical system.
Curiously, the roughly estimated amount of time required for the programs proposed and discussed above did not exceed by a significant amount the total MIPS GTO allocation. Thus, we still have no need to cut programs.
It appears that the next task is to identify with finality just what sources and survey fields we would place at the highest priority. This choice must depend not only on the scientific value of a given source/field, but also on a consideration of infrared cirrus, efficiency of mapping with SIRTF, etc. Once we have such lists, we can begin to accumulate the necessary preparatory observations to support the GTO program. In addition, a specific list of sources gives us the basis to talk more seriously about collaborations with other members of SIRTF, as well as outside collaborators.
At the same time, maybe our imaginations need more exercise than they have
gotten over the last 15 years we've worked on SIRTF. I find it surprising
that we still haven't succeeded in coveting vastly more GTO time than we
have! If this summary leads to new ideas or expansions of old ones, all
the better.
To Top