To the Level 1 BCD level, our pipelines yield quantitatively similar results to those described in Gordon et al. (2005, PASP, 117, 503). For best results, the Level 2 post-BCD pipeline (e.g. mosaic construction) needs to be customized by each user for the specific observation and the science goals; see below.
The details of the MIPS-Ge data reduction are described in the Gordon et al. (2005) paper. The two main pipeline processing steps that produce the Basic Calibrated Data (BCD) product are (1) the calculation of the slopes of the data ramps and (2) the calibration of the slope images. Also see Figure 5.6 for a flowchart of the MIPS Ge (70 and 160 µm) pipeline.
Slope Calculation: The RAW data cubes for each DCE are reduced to an uncalibrated slope image by calculating the slopes of the data ramps for each pixel. The ability to measure the slopes is affected by the high rate of cosmic ray hits on the detectors. A Bayesian technique (Hesselroth et al., 2000, Proc. SPIE Int. Soc. Opt. Eng., 4131, 26) is used to identify statistically significant changes in the slope of the ramp associated with cosmic ray events. Slopes are calculated on the ramp segments between cosmic ray events and are averaged to yield the final slope for each pixel.
Slope Calibration: After the calculation of the slopes, the uncalibrated slope image is calibrated to produce the BCD (see section 6 of Gordon et al. 2005). The key aspect of the calibration of the MIPS-Ge arrays is the frequent use of stimulator flashes to track the responsivity variations of the Ge detectors as a function of time. To accurately measure the amplitude of the stimflash requires a background DCE to be observed at the same position on the sky (and immediately before) the stimflash DCE. The stimflash-background signal is interpolated as a function of time to derive the stimflash response function. The stimflash response function is divided from the slope images to remove the time dependent responsivity variation from the data. The dark correction is then subtracted from the data, and the data are divided by a normalized illumination correction, which removes combined effect of the pixel-to-pixel gain variations and stimflash illumination pattern. The data are flux-calibrated by applying a flux conversion factor derived from observations of calibrator sources to put the BCDs in units of MJy/sr (surface brightness).
The goal of BCD pipeline processing is to produce the best possible data products that can be derived from automated processing. The pipeline produces two BCD products: (1) *_bcd.fits which is the standard calibrated BCD and (2) *_fbcd.fits which is a 'filtered' bcd product designed for point sources. The fbcd is produced by subtracting off a median of the surrounding DCEs as a function of time per pixel. This filtering technique significantly mitigates the accumulation of stimflash latents and the residual background drifts due to variations of the slow response as a function of time. The application of the median filter removes the background from both the sky and the residual detector effects (i.e., there is a loss of information about the extended background level in the field). Tests show that the application of a median filter maintains point source calibration for scan maps of fields with uniform backgrounds, but the fbcds do not preserve calibration for extended sources or for bright (>~0.2 Jy) point sources within complex emission regions. For extended and/or bright sources, you may achieve better results with offline custom filtering using the GeRT (see section 8.2.3).
Figure 5.6: MIPS Ge (70 and 160 µm) science pipeline.