6 Observations strategyWe propose to map regions of 3 square degrees on average toward 16 targets (see Table1) with IRAC (3.83 hr per square degrees) and MIPS (2.13 hr per square degrees). We will also obtain spectral maps toward 36 individal regions (including 9 reference positions) with IRS in the low resolution mode (1.3 hr per position) and MIPS in the SED mode (0.75 hr per position). Details about the individual AORs used is given in the following subsections. The total requested observing time is 646 hrs. 6.1 IRAC and MIPS imagingFor the MIPS and IRAC images, our observing strategy is defined so as to optimize the sensitivity and the photometric accuracy of the diffuse infrared emission on all angular scales. Redundancy is needed in order to correct for long-term detector response changes (see Sec.7.1.2). Our basic observing tile is a map of !1 square degrees (size constrained by the maximum duration of AORs) with in-scan and cross-scan overlap. Such a raster naturally provides redundancy over a range of timescales. The tile will be repeated with a time delay of about 4 days in order to allow a slight rotation of the scan direction between the two visits. The same observing strategy is adopted for all objects ensuring data base homogeneity. The maps will be on average 3 square degrees in size. They will be obtained by mosaicing individual tiles on the sky. The area of each map is set by the need to extend the mapping area to reference regions outside of the target. The basic IRAC observing tile will consist of a 24x11 raster and a step size of 292.8"x153.6" (40 pixels and half an array overlap in the in-scan and cross-scan directions, respectively) leading to a total area of 1.96x0.51 degrees covered with a redundancy factor of 2. At each raster position, we will obtain a single image with the 30 s integration time. The major axis of the raster will be oriented along the array coordinates (i.e. in the satellite reference frame). The AOR includes shorter integration time images for higher dynamics. The total time for this AOR is 3.83 hr with an observing efficiency of 67%. This approach brings redundancy on different time scales for each sky pixel, therefore maximizing the possibilities of characterizing detector drifts. In addition, the redundancy provided by the independent maps can be used to quantify photometric uncertainties in the final sky maps as a function of angular scales. The MIPS observations will be performed in the scan map mode using the fastest scan speed available (17 arcsec/s) and scan length of 2 degrees. A cross-scan separation of 148 arcsec (1/2 the array width) will be used, providing a full coverage of the sky at 160 micron and additional redundancy at 70 and 24 micron. Using this observing strategy, an area of 1 square degrees (13 legs long) can be mapped in 2.13 hrs, within the 3 hrs maximum time available for one MIPS AOR. Regions that might saturate the MIPS detectors, in particular at 160 micron (a prohibitive problem for dust studies in the Galactic plane) have been avoided on the basis of the IRAS 100 micron maps and the dust temperature expected for each object. 6.2 Spectroscopic ObservationsThe purpose of the spectroscopic observations is to understand the origin of the color variations that will be seen in the large scale maps. The positions to be observed will be selected from the SEDI SIRTF maps. The spectroscopic observation will be performed with the second generation AOTs, in the second year of the SIRTF schedule.
IRS low resolution spectroscopyThe IRS observations will make use of the spectral mapping AOR with no peakup. For the Long and Short slits, we will program a scan map with an inter-leg shift equal to the separation between the two orders. These two maps are designed so that a common sky area of 0.9'x2.4' is observed at all IRS wavelengths. The size and shape of this area is set by the length of the slits and their relative angle. The two independent AORs will be constrained to occur within two days so that the relative orientation of the two slits on the sky remains close to its value on the focal plane. The time constraint is also required to minimize variations in the zodiacal background emission. For each cloud, we will also observe a reference position outside the cloud in order to measure the Zodiacal Light emission and dark current levels. This reference position will be observed in sequence with the on-cloud spectra. With an exposure time of 30 s for the long wavelength part of the spectrum and 14 s for the short wavelength part, the total observing time per spectrum is 1.3 hrs, leading to an effective 1.7 hrs since we plan to include one reference for every three on-cloud spectra. The brightness sensitivity achieved in this spectral imaging AOR is reported in Fig 1. This curve does not take into account systematic errors introduced by variations in the detector reponse between the cloud and reference positions. At long wavelengths, the sensitivity is likely to be limited by these variations. We will use the IOC IRS data to model the detector response as a function of illumination. We will select the sources for second-look IRS observations (to be perfomed 1 yr after end of IOC) on the basis of the accuracy to which we will be able to correct for these variations. An accuracy of ~1% will be necessary in order to obtain a 14-40 micron spectrum of a faint translucent cloud heated only by the solar neighborhood interstellar radiation field, such as in Chamaeleon. If we cannot achieve this accuracy, the sources to be observed with IRS will be selected within the list of brighter clouds heated by associated stars.
MIPS SEDThe MIPS low resolution grating spectrometer offers the opportunity to determine the spectral energy distribution with R~15-25 in the wavelength range 55-96 micron. Our basic observing mode for MIPS SED will cover the same area as that covered with our IRS scan map. The MIPS SED slit is parallel to, and longer than, the longest side of IRS spectral imaging field. Thus, with three pointings displaced by 0.3 arcmin we can cover the full area of the IRS observations. In most cases, the chopping distance (<3 arcmin) will be too small for proper background subtraction. For each pointing we will use an observing unit of 5 cycles of 10 s. As for IRS, we will thus include a reference position selected from the MIPS 70 micron maps. The total observing time for 3 on-cloud spectra and 1 reference spectrum is 3.0 hr. The corresponding sensitivity at 38 arcsec resolution (Fig 1) provides a S/N = 10 for a cirrus cloud with Av= 0.2 mag. Within the present options on the beam throw (1, 2 or 3 arcmin) the off positions fall outside of our target spectroscopic region. If it is possible in future upgrades of the MIPS SED AOT, we will use smaller beam throws, which will improve the efficiency of these observations.
|