2. Observations and Data Reduction

Simultaneous 450 and 850 µm continuum photometry observations of 90 YSOs in the Taurus-Auriga star-forming region were obtained with the Submillimeter Common User Bolometer Array (SCUBA: Holland et al. 1999) at the 15m James Clerk Maxwell Telescope (JCMT) between 2004 February and 2005 January. Accurate reference coordinates (to <122) for each object were obtained from the 2MASS Point Source Catalog. The effective FWHM beam diameters for SCUBA photometry are 922 and 1522 at 450 ({tex}\lambda_{eff}{/tex} = 443 µm) and 850 µm ({tex}\lambda_{eff}{/tex} = 863 µm), respectively. The precipitable water vapor (PWV) levels in these observations were 1.6mm in the mean, corresponding to zenith opacities of 0.32 at 850 µm and 1.73 at 450 µm. More than 50% of the observations were conducted in very dry conditions (PWV {tex}\leq{/tex} 1.5mm). The data were acquired in sets of 18 s integrations in a small nine-point jiggle pattern with the secondary mirror chopping (typically) 6022 in azimuth at 7.8Hz. Each set consisted of between 15 and 40 integrations, and each source was usually observed for two sets. Frequent skydip observations were used to determine atmospheric extinction as a function of elevation and time. Pointing updates on nearby bright standard sources were conducted between sets of integrations: the rms pointing offsets were {tex}\leq{/tex} 222. Mars and Uranus were used as primary flux calibrators, observed at least once per night when available. The secondary calibrators HL Tau, CRL 618, and CRL 2688 were also observed approximately once every 60 to 90 minutes.

The demodulated SCUBA data were flatfielded, despiked, and corrected for extinction and residual sky emission using standard tasks in the SURF software package (Jenness & Lightfoot 1998; Jenness, Lightfoot, & Holland 1998). The "unused" bolometers in the SCUBA arrays provide a distinct advantage in sky subtraction over the standard simple demodulation utilized for single-element (or small array) detectors. With SCUBA, this technique has resulted in a factor of <3 increase in the signal-to-noise ratio (Holland et al. 1999) and should give more robust flux measurements. The mean and standard deviation voltages were used to determine the flux density and rms noise level for each source, after appropriate scaling based on the gain values derived from observations of the calibrators. Repeated observations of the flux calibrators in a given night of observing indicate a systematic uncertainty in these gain factors of <10% at 850 µm and <25% at 450 µm. These systematic errors dominate the uncertainties for brighter sources. Observations of an additional 44 sources were obtained from the SCUBA online archive and reduced in the same manner, accounting for the differences in filter sets for data taken before 1999 November. In a few cases, observations of the same source from several different nights were combined after the processing to yield very sensitive data. For those cases, the combined data are consistent with the individual datasets when the increased integration time is considered.

The Submillimeter High Angular Resolution Camera (SHARC-II: Dowell et al. 2003) on the 10m Caltech Submillimeter Observatory (CSO) telescope was also used to image 39 YSOs in the 350 µm continuum between 2004 March and 2005 January. The FWHM beam diameter of the SHARC-II point-spread function at 350 µm is roughly 922, achieved by employing an active dish surface optimization system at the CSO. Due to the low atmospheric transmission at this wavelength, these observations were only conducted when the PWV level was {tex}\leq{/tex} 1.6mm, corresponding to 350 µm zenith opacities of less than <2. Opacity measurements at 225GHz were taken every 10 minutes with a dedicated tilting water vapor monitor observing at fixed azimuth. The observations were conducted by constantly sweeping the telescope in the vicinity of the source in an alt-az Lissajous pattern, providing small Nyquist-sampled maps. At least three separate maps were taken for each source, with between 120 and 600 s of integration per map. The aforementioned SCUBA calibrators were also observed every 60 to 90 minutes for pointing updates and flux calibration. The SHARC-II data reduction was conducted using the CRUSH software package (Kov´acs et al. 2005). Flux densities were measured in a circular aperture with a radius of 3022, and rms noise levels were determined from the background pixels. Repeated measurements of standard calibration sources show that the absolute flux calibration is accurate to within 25%.

Our sample was selected primarily from the compilation of Kenyon & Hartmann (1995), and was designed to contain roughly equal numbers of Class II and III objects, WTTSs and CTTSs, and single and multiple stars. The histograms in Figure 1 summarize some of the key properties of the sample. Table 1 gives a collection of submillimeter properties for 153 YSOs in Taurus-Auriga: 90 sources with new SCUBA and SHARC-II data, 44 with archival SCUBA observations and SHARCII data, 4 with data from the literature, and 15 others with SHARC-II data and flux densities from the literature (see the table notes). This table lists the 350, 450, 850 µm, and 1.3mm flux densities (the latter from the literature) and statistical errors ({tex}1-\sigma{/tex} rms noise levels) or {tex}3-\sigma{/tex} upper limits in units of mJy per beam, disk masses (see §3.2) and submillimeter continuum slopes (see §3.3), and various other relevant properties. The projected FWHM beam diameters at the assumed distance of Taurus-Auriga (d = 140 pc; Elias 1978) are 1260AU for both 350 and 450 µm and 2100AU for 850 µm. In this paper, we assume that all of the sources are unresolved, and therefore the values in Table 1 are actually the integrated continuum flux densities (in units of mJy). This assumption is valid for Class II and III sources, where the submillimeter emission originates in a disk with a radius of a few hundred AU at most (see the interferometric observations of Dutrey et al. 1996; Kitamura et al. 2002).

On the other hand, submillimeter continuum maps of the Class I sources in this sample usually show a significant amount of extended emission from the outer envelope in addition to a bright, central concentration of emission (itself perhaps marginally resolved) from the disk and inner envelope (e.g., Chandler & Richer 2000; Hogerheijde & Sandell 2000; Shirley et al. 2000; Motte & André 2001; Chini et al. 2001; Young et al. 2003). The non-mapping photometry observations at 450 and 850 µm presented here exclude the extended emission component, and therefore only sample the bright peak of emission which presumably originates from warm dust in the inner envelope and/or a disk. Because of the unknown density structure of the inner envelope, it is not possible to unambiguously determine what fraction of this emission peak is contributed by a compact object (i.e., disk) without interferometric observations (see the discussion by Young et al. 2003, and references therein). For some of these objects, there is also the possibility that the 6022 chop throw would place the "off" position in the extended envelope emission, and therefore the flux densities listed in Table 1 could be slightly underestimated. The reader should keep in mind that the submillimeter properties of Class I YSOs in this paper most likely refer to a combination of disk and envelope contributions. The notes in Table 1 provide references to submillimeter maps of the Class I YSOs in the literature when available.

The primary observational goal of this survey was to take advantage of the stability and efficiency of the SCUBA instrument to obtain a 850 µm sample with a relatively uniform flux density limit of <10mJy ({tex}3-\sigma{/tex}). This means {tex}3-\sigma{/tex} upper limit for undetected sources at 850 µm in this survey is 8.4mJy (the median is the same), with a standard deviation in the upper limits of 3.1mJy. For comparison, the same sources in the combined 1.3mm surveys in Taurus-Auriga conducted by Beckwith et al. (1990) and Osterloh & Beckwith (1995) have a mean {tex}3-\sigma{/tex} upper limit of 19mJy (median of 16mJy) and a standard deviation in the upper limits of 10mJy. If we assume that the submillimeter continuum emission behaves as {tex}F_v \propto v^2{/tex} (see §3.3), then a factor of 2.3 can be used to scale the 1.3mm measurements with those at 850 µm. The resulting scaled 1.3mm mean upper limit is then 44mJy (median of 37mJy). The distributions of the upper limits of undetected sources are shown in Figure 2. In terms of flux density limits on undetected sources, our survey is roughly a factor of 5 more sensitive than previous single-dish work and is also considerably more uniform. The distributions of the signal-to-noise ratios for detected sources in the various surveys are similar, although there are generally higher ratios at 850 µm. For the sources common to the 850 µm and 1.3mm samples, the detection rates are 64 ± 7% and 47 ± 6%, respectively.