This data set description was provided by Buie and McLaughlin.  Data Set Overview:  This data set contains broadband R images and derived photometry of comet 9P/Tempel 1, the target of the Deep Impact mission. Data were acquired by Buie at the Perkins 72-inch telescope of the Lowell Observatory during 11 nights of observing from 28 September 2000 through 14 January 2001, about 8.5 to 12.5 months after the comet passed through perihelion on 2 January 2000.  The observations were made using three different, visible CCD cameras mounted at the telescope:  Observing Dates Detector Used Plate Scale --------------- ----------------- ---------------------------- 2000 Sep-Oct Loral 2K x 2K* 0.726 arcsecond/binned pixel 2000 Nov-Dec SITe 2K x 2K 1.134 arcsecond/binned pixel 2001 Jan Navy TI 800 x 800 0.800 arcsecond/binned pixel  * Buie noted the Loral CCD unbonded from its mount sometime in late October 2000 but was repaired.  This data set also includes dark frames and broadband BVR sky flats and Landolt standard star fields used to calibrate the raw comet images. Photometric and lightcurve results were presented by Meech, et al. (2000) [MEECHETAL2000].  Parameters:  Observational parameters for each image are incorporated into the PDS labels and FITS headers supplied with the data.  Processing:  Buie developed and, with Pate, used IDL routines to de-bias, flatten, and calibrate the raw images. Several Landolt star fields (Landolt, 1992 [LANDOLT1992]) were used to calibrate the comet images. Daily finder charts and lists that identify the calibration are included as documentation for this data set.  Buie noted a special bias correction step was applied to data taken from September through October 2000 with the Loral CCD. The CCD, prior to coming unbound from its mount in late October, experienced a variable bias. Buie developed a method to remove or at least minimize its effect on these data:  By plotting bias level versus time through the night, we could see that the bias level would jump by a discrete and nearly constant level for 10's of images. By averaging these apparent sets we determined a group correction so that the entire night's data kept a reasonable bias level.  The bias correction table that was used during the reduction process is included in the calibration directory. Images taken with the SITe and NAVY CCDs did not require this special bias correction because both CCDs had a stable. A standard bias subtraction was applied to these data.  For each night, Buie performed differential photometry of the calibrated R images and produced lightcurves in the highest signal-to-noise ratio aperture. Daily lightcurve data are provided as photometry data tables. Output from multiple-size aperture and single aperture (best) photometry extractions are provided as data tables as well as documentation for this data set.  Data:  Images :  Raw and reduced images are supplied as FITS files with headers produced by telescope software and IDL processing, with detached PDS labels generated by the Applied Coherent Technology Corporation and the Small Bodies Node. In the FITS image array, North is down and East is to the right. This orientation is specified by the LINE_DISPLAY_DIRECTION and SAMPLE_DISPLAY_DIRECTION keywords in the PDS labels.  This data set contains 439 raw and reduced images of the comet, 698 raw sky flats, darks, standard star fields, and 268 reduced standard star fields. On 2000-12-16, 17 images of Messier objects (M1, M31, M42, M78, M79, and IC 433) were taken but not reduced. On night 2000-12-16, one test frame was taken and on 2000-10-17, one focus frame was taken. These frames are given a target name of 'N/A' in the PDS labels.  The following table provides the date and UT time range for comet images taken during the 11 observing nights. Guiding was performed on the comet. An estimate of the amount of trailing for stars, in arcseconds, was obtained from the image and provided in the following table. The sky conditions in the following table were taken from the observing logs in the document directory.  #Comet Imgs Trail Obs Date UT Time by Filter ('') Sky Condition ---------- ----------- ----------- ---- -------------------------- 2000-09-28 07:11-09:41 20 R 6-9 Cummulus then clear 2000-09-29 03:43-09:38 53 R, 2 V 6-9 Clearing, some thin clouds 2000-09-30 04:27-09:36 54 R 6-9 Clearing 2000-10-01 05:08-09:23 38 R 6-9 Clearing late 2000-10-17 02:25-08:11 56 R 5-7 Smoke 2000-10-18 02:27-08:02 40 R, 5 U* 5-7 Clear 2000-11-19 01:41-06:06 40 R 2-3 Clear 2000-11-20 01:34-06:06 48 R 2-3 Clear 2000-12-16 01:10-04:35 39 R 4-5 Cirrus 2000-12-18 01:11-04:11 29 R, 1 V 4-5 Clearing 2001-01-14 02:03-03:35 14 R 2-3 Some clouds low on horizon  * Broadband U images were taken by accident.  The following table provides the geometric circumstances of the comet for the 11 observing nights. The values were generated by the JPL Horizons System. The last column in the  Phase Heliocent. Geocent. Angle Obs Date dist. [AU] dist. [AU] [deg.] ---------- ---------- ---------- ------ 2000-09-28 2.760 1.794 6.9 2000-09-29 2.765 1.803 7.1 2000-09-30 2.771 1.811 7.3 2000-10-01 2.777 1.820 7.6 2000-10-17 2.864 2.000 11.9 2000-10-18 2.870 2.013 12.1 2000-11-19 3.040 2.528 17.5 2000-11-20 3.045 2.546 17.6 2000-12-16 3.176 3.042 18.0 2000-12-18 3.186 3.080 18.0 2001-01-14 3.317 3.593 15.7  Photometry :  Buie provided three types of photometry extractions:  - Differential photometry for each night for lightcurve analysis.  - Single aperture extractions for each night at an aperture that gave the best possible signal-to-noise ratio. The aperture was matched to the seeing for that night, regardless of the scale of the image at the comet.  - Multiple aperture size extractions for each night. Each nightly set of apertures ranged from a minimum size determined by the maximum signal-to-noise ratio and less than 1% photometric error up to 10 arcseconds in 1 arcsecond increments above 5 arcseconds.  Media/Format:  The data are supplied as an electronic file in a PDS standard volume structure.

Confidence Level Overview :  Standard techniques were used for the reduction of the image data and calculation of the photometry.  Some data frames have ghost images caused by filter reflections.  Lightcurve photometry is presented as differential R magnitudes with error bars. However, it appears that the R magnitudes are apparent magnitudes and are not corrected to absolute magnitudes, that is R(1,1,0).  Review :  This data set was reviewed during a PDS Peer review held in May 2004.