+------------------------------------------------------------------------------+ New HST Multi-Wavelength Imaging of the Eagle Nebula 05 January 2014 Jennifer Mack (mack@stsci.edu) Research & Instrument Scientist Space Telescope Science Institute 3700 San Martin Drive, Baltimore, MD 21218 Table of Contents: 1. Introduction and observations 2. New drizzling software 3. Image registration and combination 4. Improving the UVIS calibration 5. Data products 6. Citation +------------------------------------------------------------------------------+ 1. Introduction and observations One of the most iconic images from the Hubble Space Telescope has been the 1995 WFPC2 image of the Eagle Nebula, also known as the 'Pillars of Creation'. (For details, see Proposal 5773 and news release http://hubblesite.org/newscenter/archive/releases/1995/44/). Nineteen years after those original observations, new mosaics have been obtained with HST's Wide Field Camera 3 using both the UVIS and IR channels. The wider field of view, higher resolution, and broader wavelength coverage of the new images highlight the improved capabilities of HST over its long-lasting operation, made possible by the upgraded instrumentation installed during Space Shuttle Servicing Mission 4 in 2009. The new observations were obtained in September 2014 in 20 orbits and include a small IR mosaic and a larger UVIS mosaic. A summary of the WFC3 filters is given in Table 1. (See Proposal 13926 http://www.stsci.edu/hst/phase2-public/13926.pro, PI: Levay for details.) The mosaic tiles were generated using the APT mosaicking tool with the IR target centered at RA= 18H 18M 52.1296S, DEC=-13D 50' 9.80" (J2000) in the IR-FIX aperture and the UVIS target centered at RA=18H 18M 50.5000S, DEC=-13D 49' 44.00" (J2000) in the UVIS1-FIX aperture. The tile (visit) positions are labeled in Figure 1, where the HST file naming convention gives the visit number in the 5th and 6th characters. For example, the archival data product ick901020_drz corresponds to tile 01. Note that for the UVIS detector, visits 09 and 10 comprise a 2x1 mosaic which overlaps the primary 2x2 UVIS mosaic (Visits 05-08). These 2 additional visits were added obtain greater signal-to-noise in the pillars. As part of the observing strategy, small shifts (dithers) between tiles have been obtained to allow for the removal of detector artifacts such as IR blobs and the UVIS chip gap. These 'fine' dither patterns for the UVIS and IR mosaics are summarized in Table 2. The total exposure time and number of images used to create the five filter mosaics is given in Table 3. A summary the individual IR and UVIS frames are given in Tables 4 and 5, respectively, and include the association name, the rootname of the individual FLT exposures making up the association, the filter, date of observation, the sample sequence and number of samples (IR only), total exposure time, and the X and Y dither (POSTARG) between tiles given in arcseconds. Color composite images will be presented at the 225th meeting of the AAS to commemorate the 25th anniversary of HST's launch. The carefully calibrated, aligned, and combined drizzled data products used to make these mosaics are now available as High-Level Science Products in MAST (http://archive.stsci.edu/prepds/heritage/m16). Parallel observations with the ACS Wide Field Channel create an H-alpha mosaic just north of the target, and HLSPs for this dataset will provided in a subsequent release. Table 1: Filters ---------------------------- Detector Filter Description Pivot_Wave(nm) Rect_Width(nm) Zeropoint* --------- ----- -------------------- ------ ----- -------- WFC3/UVIS F502N OIII [5007] 501.0 6.5 22.0013 WFC3/UVIS F657N Wide H-alpha + [NII] 656.7 12.1 22.9310 WFC3/UVIS F673N [SII] 6717/6731 676.6 11.8 22.9228 WFC3/IR F110W Wide Y 1153.4 443.0 28.2602 WFC3/IR F160W WFC3 H 1536.9 268.3 27.9963 *Note that the WFC3 STMAG zeropoints in column 6 are given for an aperture 0.4". For more details (including any updates to the values), please see the WFC3 zeropoints webpage http://www.stsci.edu/hst/wfc3/phot_zp_lbn Figure 1: APT Mosaics ---------------------------- In the diagrams below, the mosaic tiles are labeled by visit (tile) number, which can be found in the 5th & 6th characters of the HST filename. ____ ____ | | | IR Mosaic 2x2 2 dithers per filter F110W, F160W = 1 orbit/tile = 4 orbits | 02 | 01 | |____|____| | | | | 04 | 03 | |____|____| _______ _______ | | | UVIS Mosaic 2x2 3 dithers per filter F502N, F657N, F673N = 3 orbits/tile = 12 orbits | | | | 06 | 05 | |_______|_______| | | | | 08 | 07 | | | | |_______|_______| _______ | | UVIS Mosaic 2x1 2 dithers per filter F502N, F657N, F673N = 2 orbits/tile = 4 orbits | | | 09 | |_______| | | | 10 | | | |_______| Table 2: Fine Dither Patterns ------------------------------ Pattern_Type WFC3-IR-DITHER-BLOB Pattern_Purpose DITHER Number_Of_Points 2 Point_Spacing 7.2 arcsec Pattern_Orient 90 Center_Pattern YES Pattern_Type WFC3-UVIS-MOSAIC-LINE Pattern_Purpose MOSAIC Number_Of_Points 3 Point_Spacing 12 arcsec Pattern_Orient 65 Center_Pattern NO Table 3: Combined Data Products -------------------------------- IMAGE FILTER EXPTIME N_Images N_Tiles hlsp_heritage_hst_wfc3-uvis_m16_f502n_v1_drz.fits F502N 16000.0 sec 16 6 hlsp_heritage_hst_wfc3-uvis_m16_f657n_v1_drz.fits F657N 9600.0 sec 16 6 hlsp_heritage_hst_wfc3-uvis_m16_f673n_v1_drz.fits F673N 14400.0 sec 16 6 hlsp_heritage_hst_wfc3-ir_m16_f110w_v1_drz.fits F110W 4423.5 sec 8 4 hlsp_heritage_hst_wfc3-ir_m16_f160w_v1_drz.fits F160W 5623.5 sec 8 4 Table 4: WFC3/IR observations ASSOCIATION ROOTNAME FILTER DATE SAMP_SEQ NSAMP EXPTIME POSTARG1 POSTARG2 --------------------------------------------------------------------------------------------------------------------------------------- ICK901020 ick901hxq F110W 2014-09-01 SPARS50 13 552.936890 -62.566380 -62.047470 ICK901020 ick901i4q F110W 2014-09-02 SPARS50 13 552.936890 -62.566380 -54.847469 ICK902020 ick902n7q F110W 2014-09-02 SPARS50 13 552.936890 62.566380 -62.047470 ICK902020 ick902nbq F110W 2014-09-02 SPARS50 13 552.936890 62.566380 -54.847469 ICK903020 ick903n2q F110W 2014-09-07 SPARS50 13 552.936890 -62.566380 54.847469 ICK903020 ick903n9q F110W 2014-09-07 SPARS50 13 552.936890 -62.566380 62.047470 ICK904020 ick904o9q F110W 2014-09-07 SPARS50 13 552.936890 62.566380 54.847469 ICK904020 ick904odq F110W 2014-09-07 SPARS50 13 552.936890 62.566380 62.047470 ICK901030 ick901hzq F160W 2014-09-02 SPARS50 16 702.938171 -62.566380 -62.047470 ICK901030 ick901i7q F160W 2014-09-02 SPARS50 16 702.938171 -62.566380 -54.847469 ICK902030 ick902n9q F160W 2014-09-02 SPARS50 16 702.938171 62.566380 -62.047470 ICK902030 ick902neq F160W 2014-09-02 SPARS50 16 702.938171 62.566380 -54.847469 ICK903030 ick903n4q F160W 2014-09-07 SPARS50 16 702.938171 -62.566380 54.847469 ICK903030 ick903ncq F160W 2014-09-07 SPARS50 16 702.938171 -62.566380 62.047470 ICK904030 ick904obq F160W 2014-09-07 SPARS50 16 702.938171 62.566380 54.847469 ICK904030 ick904ogq F160W 2014-09-07 SPARS50 16 702.938171 62.566380 62.047470 Table 5: WFC3/UVIS observations ASSOCIATION ROOTNAME FILTER DATE EXPTIME POSTARG1 POSTARG2 --------------------------------------------------------------------------------------------------------------------------------------- ICK905030 ick905k3q F502N 2014-09-02 1000.000000 -65.383713 -74.272148 ICK905030 ick905kaq F502N 2014-09-02 1000.000000 -60.312290 -63.396450 ICK905030 ick905kjq F502N 2014-09-02 1000.000000 -55.240871 -52.520760 ICK906030 ick906kuq F502N 2014-09-02 1000.000000 65.383713 -65.127998 ICK906030 ick906l1q F502N 2014-09-02 1000.000000 70.455132 -54.252300 ICK906030 ick906laq F502N 2014-09-02 1000.000000 75.526550 -43.376610 ICK907030 ick907niq F502N 2014-09-02 1000.000000 -65.383713 65.127998 ICK907030 ick907nxq F502N 2014-09-03 1000.000000 -60.312290 76.003693 ICK907030 ick907oqq F502N 2014-09-03 1000.000000 -55.240871 86.879379 ICK908030 ick908p9q F502N 2014-09-03 1000.000000 65.383713 74.272148 ICK908030 ick908pgq F502N 2014-09-03 1000.000000 70.455132 85.147842 ICK908030 ick908ppq F502N 2014-09-03 1000.000000 75.526550 96.023529 ICK909030 ick909c1q F502N 2014-09-05 1000.000000 0.000000 -69.700073 ICK909030 ick909cwq F502N 2014-09-05 1000.000000 5.071420 -58.824379 ICK910030 ick910d7q F502N 2014-09-05 1000.000000 0.000000 69.700073 ICK910030 ick910deq F502N 2014-09-05 1000.000000 5.071420 80.575768 ICK905040 ick905k5q F657N 2014-09-02 600.000000 -65.383713 -74.272148 ICK905040 ick905keq F657N 2014-09-02 600.000000 -60.312290 -63.396450 ICK905040 ick905knq F657N 2014-09-02 600.000000 -55.240871 -52.520760 ICK906040 ick906kwq F657N 2014-09-02 600.000000 65.383713 -65.127998 ICK906040 ick906l5q F657N 2014-09-02 600.000000 70.455132 -54.252300 ICK906040 ick906leq F657N 2014-09-02 600.000000 75.526550 -43.376610 ICK907040 ick907nkq F657N 2014-09-03 600.000000 -65.383713 65.127998 ICK907040 ick907o1q F657N 2014-09-03 600.000000 -60.312290 76.003693 ICK907040 ick907ouq F657N 2014-09-03 600.000000 -55.240871 86.879379 ICK908040 ick908pbq F657N 2014-09-03 600.000000 65.383713 74.272148 ICK908040 ick908pkq F657N 2014-09-03 600.000000 70.455132 85.147842 ICK908040 ick908ptq F657N 2014-09-03 600.000000 75.526550 96.023529 ICK909040 ick909c3q F657N 2014-09-05 600.000000 0.000000 -69.700073 ICK909040 ick909d0q F657N 2014-09-05 600.000000 5.071420 -58.824379 ICK910040 ick910d9q F657N 2014-09-05 600.000000 0.000000 69.700073 ICK910040 ick910diq F657N 2014-09-05 600.000000 5.071420 80.575768 ICK905050 ick905k8q F673N 2014-09-02 900.000000 -65.383713 -74.272148 ICK905050 ick905khq F673N 2014-09-02 900.000000 -60.312290 -63.396450 ICK905050 ick905kqq F673N 2014-09-02 900.000000 -55.240871 -52.520760 ICK906050 ick906kzq F673N 2014-09-02 900.000000 65.383713 -65.127998 ICK906050 ick906l8q F673N 2014-09-02 900.000000 70.455132 -54.252300 ICK906050 ick906lhq F673N 2014-09-02 900.000000 75.526550 -43.376610 ICK907050 ick907nnq F673N 2014-09-03 900.000000 -65.383713 65.127998 ICK907050 ick907o4q F673N 2014-09-03 900.000000 -60.312290 76.003693 ICK907050 ick907oxq F673N 2014-09-03 900.000000 -55.240871 86.879379 ICK908050 ick908peq F673N 2014-09-03 900.000000 65.383713 74.272148 ICK908050 ick908pnq F673N 2014-09-03 900.000000 70.455132 85.147842 ICK908050 ick908pwq F673N 2014-09-03 900.000000 75.526550 96.023529 ICK909050 ick909c6q F673N 2014-09-05 900.000000 0.000000 -69.700073 ICK909050 ick909d3q F673N 2014-09-05 900.000000 5.071420 -58.824379 ICK910050 ick910dcq F673N 2014-09-05 900.000000 0.000000 69.700073 ICK910050 ick910dlq F673N 2014-09-05 900.000000 5.071420 80.575768 +--------+---------+---------+---------+---------+---------+---------+---------+ 2. New Drizzling Software This example makes use of a new 'development version' of the DrizzlePac Software (version 2.0) which has added features for aligning and combining mosaics. This software is not yet in use by the OPUS pipeline and is currently only available for download via SSBX at the following link. http://ssb.stsci.edu/ssb_software.shtml One useful new feature in TweakReg is support for aligning frames that cover a large area on the sky. When the user does not supply a reference image, TweakReg selects two images from the input list with the largest overlap and tries to align them by matching sources. One image will be used to produce the "reference catalog" while the other image will be aligned to this catalog. Now, the sources from the second image that have not been matched to the reference catalog are considered as "good new sources" and will be added to the reference catalog. This way, the reference catalog keeps expanding with each new matched image. With a large (expanded) reference catalog now it is possible to align images that had no direct overlap with the starting image. The new TweakReg also makes it easier to align different HST filters/detectors. This is now achieved in a single step by allowing for separate sets of source detection parameters (threshold, conv_width, etc) for the input image and the reference image. For example, we demonstrate how to align a set of distorted IR input frames (F110W) to a drizzled IR reference frame (F160W) and how to align a set of drizzled UVIS input frames (F657N) to a drizzled IR reference frame (F160W). AstroDrizzle also features new options for matching the sky background when tiling together large mosaics. In prior versions of the software, the sky background was based on clipped statistics (defined by the parameters skyclip, skystat, skylsigma, and skyusigma) in the images separately. This is performed for all chips and the lowest sky value (in electrons/arcs**2) among chips is then adopted. For observations of sparse fields, this approach works well. When large extended objects fill the detector, however, there is no true 'blank sky' and the background value is an overestimate. Additionally, when extended targets are observed as mosaics (eg. with large dithers), the 'scene' can change significantly between exposures and bias the background estimate. An error in determining the sky background may in turn impact the accuracy of the cosmic ray rejection, and if severe enough the resulting photometry. Additionally, by not properly matching the sky background before combining frames, correlated noise may be added to the final drizzled products when differences in the background levels are significant. Correlated noise looks like a faint 'screen door' pattern superimposed on the image. In these cases, the best approach is to give AstroDrizzle an ASCII file ('skyfile') containing user-defined background values. In the new software DrizzlePac, AstroDrizzle now features several options for computing the sky. One of these options, 'skymethod=match', is useful for “equalizing” the sky background across large mosaics. This method computes differences in sky values using only pixels in common between images. The sky values will be relative to the value computed for the input frame with the lowest sky value for which the MDRIZSKY keyword will be set to 0. For more information on the skypac.skymatch task called by AstroDrizzle, users may refer to the following webpage. http://ssb.stsci.edu/doc/stsci_python_x/stsci.skypac.doc/html/skymatch.html For more information on new software features, see the DrizzlePac webpage: http://www.stsci.edu/drizzlepac +--------+---------+---------+---------+---------+---------+---------+---------+ 3. Image registration and combination In two prior examples, we showed how to align and combine mosaics using an iterative process to build up the field of view of the reference image. Additional filters/detectors may then be aligned directly to the reference image. http://archive.stsci.edu/prepds/heritage/horsehead/readme_HLSP_v3.txt http://archive.stsci.edu/prepds/heritage/ngc2174/ With the new DrizzlePac 2.0, this iterative procedure is no longer necessary. For this large dataset, we start with the broadband WFC3/IR images which are full of stars and therefore easiest align. These may be combined to create an IR reference image with which to align the UVIS narrowband frames which are largely devoid of point sources and full of cosmic-rays. The summary below describes the process of generating the 5 filter mosaics (2 IR, 3 UVIS) and includes syntax for running each task from the command line in PyRAF. ----------------------------------- ----------------------------------- #Import the software: import drizzlepac from drizzlepac import tweakreg from drizzlepac import astrodrizzle from drizzlepac import tweakback from stwcs import updatewcs To verify which version of DrizzlePac you are running, you may type the following command. drizzlepac.__version__ '2.0.0.dev33816' In general, it is good practice to run the task 'updatewcs' ensure that the image header WCS is up-to-date and compatible with the version of the DrizzlePac software you are running before attempting to align images. updatewcs.updatewcs('i*flt.fits') ----------------------------------- ----------------------------------- #F160W: Align and combine the F160W frames. When 'expand_refcat=True' and 'enforce_user_order=False', TweakReg will select the image with the largest overlap and work out from there, expanding the reference catalog as it goes. The input list 'f160w.list' contains the names of the 8 *flt.fits frames. "f160w.list" ick901hzq_flt.fits ick901i7q_flt.fits ick903n4q_flt.fits ick903ncq_flt.fits ick904obq_flt.fits ick904ogq_flt.fits ick902n9q_flt.fits ick902neq_flt.fits tweakreg.TweakReg('@f160w.list',imagefindcfg={'threshold':50,'conv_width':2.5},expand_refcat=True,enforce_user_order=False,minobj=5, shiftfile=True,outshifts='shift160_flt.txt',sigma=2.8,searchrad=3.0,updatehdr=True,wcsname='FLT',interactive=True,ylimit=0.5) TweakReg finds between 500-600 objects per image. The contents of the shift file are given below as an example, with an additional column indicating the number of objects matched with the reference frame which was automatically selected by the task to be ick902neq_flt.fits. These reflect updates made to the header WCS required to correct for small pointing errors. The shift file is shown only for the first two IR mosaics as illustration. # image dx dy drot dscale xrms yrms #match ick902neq_flt.fits 0.000000 0.000000 0.000000 1.000000 0.000000 0.000000 -- ick902n9q_flt.fits 0.028769 0.085079 359.999794 1.000000 0.060854 0.043356 478 ick904obq_flt.fits -8.747473 -12.125906 359.997217 0.999916 0.049201 0.043300 63 ick904ogq_flt.fits -8.791378 -12.221958 359.997744 1.000070 0.061581 0.043106 384 ick901hzq_flt.fits -0.159702 0.419562 359.999331 1.000085 0.057073 0.036684 43 ick901i7q_flt.fits -0.209366 0.364715 0.000374 0.999920 0.059385 0.051056 446 ick903n4q_flt.fits -6.982110 -11.973897 359.999316 0.999995 0.065746 0.049929 56 ick903ncq_flt.fits -7.021394 -12.037023 359.999190 1.000016 0.061807 0.045871 406 Next we combine the full set of F160W images with AstroDrizzle, defining an orientation of -35 degrees such that the pillars are straight up. The final scale is set to be exactly twice that of the UVIS mosaics, and the sky background is equalized by setting the parameter 'skymethod=match'. The parameters 'driz_sep_bits' and 'final_bits' define which DQ flags in the *flt.fits[3] to treat as good. For IR data, this is typically set to 64+512 in the pipeline, corresponding to warm pixels and IR blobs. Because we performed a blob dither, however, we remove 512 from the list of good flags so that these deviant pixels are treated as bad and replaced with non-flagged pixels from the accompanying dithered frames. astrodrizzle.AstroDrizzle('@f160w.list',output='f160w_sc08px10', preserve=no, clean=no, build=no, skymethod='match', driz_sep_bits='64', driz_cr_corr=yes, final_bits='64', final_wcs=yes, final_scale=0.08, final_pixfrac=1.0, final_rot=-35, final_ra=274.721587, final_dec=-13.841549, final_outnx=4000, final_outny=4200) ----------------------------------- ----------------------------------- #F110W: Align F110W frames to the drizzled F160W reference image. The input list 'f110w.list' contains the names of the 8 *flt.fits frames. As discussed in Section 2, DrizzlePac 2.0 now makes it easier to align directly to a reference image by allowing the user to specify separate sets of source finding criteria via the imagefindcfg and refimagefindcfg parameter sets, as shown below. "f110w.list" ick901hxq_flt.fits ick901i4q_flt.fits ick902n7q_flt.fits ick902nbq_flt.fits ick903n2q_flt.fits ick903n9q_flt.fits ick904o9q_flt.fits ick904odq_flt.fits tweakreg.TweakReg('@f110w.list',imagefindcfg={'threshold':60,'conv_width':2.5},expand_refcat=False,enforce_user_order=False, refimage='f160w_sc08px10_drz_sci.fits', refimagefindcfg={'threshold':250,'conv_width':2.5}, minobj=5, shiftfile=True,outshifts='shift110_flt.txt', sigma=2.5,searchrad=3.0,updatehdr=True,wcsname='FLT',interactive=True,ylimit=0.5) With these detection thresholds, TweakReg finds between 200-300 objects in each F110W frame and ~2900 objects in the F160W reference image. Note that we set 'expand_refcat=False' for this filter since the reference catalog is fully defined by the drizzled F160W mosaic. The shift file below gives the required offsets and fit rms, where the units are now expressed in reference pixels (e.g. 0.08”/pixel versus the F160W shift file where the reference image has a scale of 0.1283”/pix). # image dx dy drot dscale xrms yrms #match ick901hxq_flt.fits -0.003019 0.471148 0.001507 1.000058 0.186245 0.133508 247 ick901i4q_flt.fits -0.043820 0.431017 0.001427 1.000071 0.183568 0.127997 230 ick902n7q_flt.fits 0.449591 -0.137317 0.000081 1.000077 0.178102 0.127555 276 ick902nbq_flt.fits 0.376117 -0.191166 0.000470 1.000058 0.170392 0.119254 283 ick903n2q_flt.fits -10.896301 -19.298041 0.000337 1.000066 0.176379 0.127778 184 ick903n9q_flt.fits -10.905232 -19.369878 0.000253 1.000073 0.163772 0.132321 194 ick904o9q_flt.fits -13.619104 -19.629480 359.998923 1.000007 0.151163 0.128454 199 ick904odq_flt.fits -13.675618 -19.685206 359.998026 1.000008 0.193882 0.129358 201 Next, we combine the full set of F110W images, where the output reference frame (eg. scale, orientation, image size, etc.) is defined by the drizzled F160W image. astrodrizzle.AstroDrizzle('@f110w.list',output='f110w_sc08px10', preserve=no,clean=no, build=no, skymethod='match', driz_sep_bits='64', driz_cr_corr=yes, final_bits='64', final_wcs=yes, final_refimage='f160w_sc08px10_drz_sci.fits') ----------------------------------- ----------------------------------- #F657N: Align UVIS F657N frames to the drizzled F160W reference image. The input list 'f657n_DRZ.list' contains the names of the 6 visit-level *drz.fits products from MAST. Rather than aligning the *flt.fits frames themselves, this alternate approach is used because the long UVIS exposures are full of cosmic-rays and few point sources. (IR frames, on the other hand, are full of point sources and have had cosmic-rays rejected by CALWF3 up-the-ramp fitting.) Exposures making up the visit-level drizzled products are typically aligned to ~2-5 milliarcseconds with fine-lock on 2 guide stars. For observations obtained in different visits using the same set of guide stars, offsets of ~50-100 milliarcseconds are typical, and tweaks to the header WCS are necessary. For visits using different sets of guide stars, the pointing accuracy is typically 0.2-0.5 arcseconds. "f657n_DRZ.list" ick905040_drz.fits ick906040_drz.fits ick907040_drz.fits ick908040_drz.fits ick909040_drz.fits ick910040_drz.fits tweakreg.TweakReg('@f657n_DRZ.list',imagefindcfg={'threshold':10,'conv_width':3.5}, refimage='f160w_sc08px10_drz_sci.fits', refimagefindcfg={'threshold':200,'conv_width':2.5},minobj=5, shiftfile=True,outshifts='shift657_drz.txt',sigma=2.5,searchrad=5.0,updatehdr=True,wcsname='DRZ',interactive=True,ylimit=0.5) Use TweakBack to propagate the updated WCS back to the *flt.fits frames making up each association. tweakback.tweakback('ick905040_drz.fits',verbose=True) tweakback.tweakback('ick906040_drz.fits',verbose=True) tweakback.tweakback('ick907040_drz.fits',verbose=True) tweakback.tweakback('ick908040_drz.fits',verbose=True) tweakback.tweakback('ick909040_drz.fits',verbose=True) tweakback.tweakback('ick910040_drz.fits',verbose=True) Drizzle the set of 16 UVIS F657N *flt.fits frames to a scale of 0.04"/pix (half the plate scale of the IR frames), using the same central RA/Dec and orientation. The IR mosaic size was set to 4000x4200 pixels and the corresponding UVIS mosaic to 8000x8400 pixels. astrodrizzle.AstroDrizzle('@f657n.list', output='f657n_sc04px10', clean=no, build=no, skymethod='match', driz_sep_bits='96', driz_cr_corr=yes, final_bits='96', final_wcs=yes, final_scale=0.04, final_pixfrac=1.0, final_rot=-35, final_ra=274.721587, final_dec=-13.841549, final_outnx=8000, final_outny=8400, num_cores=1, preserve=False) ----------------------------------- ----------------------------------- #F673N: Align UVIS F673N frames to the drizzled F657N reference image. The input list 'f673n_DRZ.list' contains the names of the 6 visit-level *drz.fits products from MAST. Because the IR mosaics subtend a smaller area on the sky than the UVIS mosaics, we chose to use the F657N drizzled product as reference rather than the F160W drizzled product to ensure that the 3 UVIS filters are aligned as accurately as possible. "f673n_DRZ.list" ick905050_drz.fits ick906050_drz.fits ick907050_drz.fits ick908050_drz.fits ick909050_drz.fits ick910050_drz.fits tweakreg.TweakReg('@f673n_DRZ.list',imagefindcfg={'threshold':10,'conv_width':3.5},refimage='f657n_sc04px10_drz_sci.fits', refimagefindcfg={'threshold':100,'conv_width':3.5},minobj=5, shiftfile=True,outshifts='shift673_drz.txt',sigma=2.5,searchrad=5.0,updatehdr=True,wcsname='DRZ',interactive=True,ylimit=0.5) from drizzlepac import tweakback tweakback.tweakback('ick905050_drz.fits',verbose=True) tweakback.tweakback('ick906050_drz.fits',verbose=True) tweakback.tweakback('ick907050_drz.fits',verbose=True) tweakback.tweakback('ick908050_drz.fits',verbose=True) tweakback.tweakback('ick909050_drz.fits',verbose=True) tweakback.tweakback('ick910050_drz.fits',verbose=True) Drizzle the set of 16 UVIS F673N *flt.fits frames using the F657N mosaic as a reference. astrodrizzle.AstroDrizzle('@f673n.list', output='f673n_sc04px10', clean=no, build=no, skymethod='match', driz_sep_bits='96', driz_cr_corr=yes, final_bits='96', final_wcs=yes, final_refimage='f657n_sc04px10_drz_sci.fits', preserve=False) ----------------------------------- ----------------------------------- #F502N: Align UVIS F502N frames to the drizzled F657N reference image. The input list 'f502n_DRZ.list' contains the names of the 6 visit-level *drz.fits products from MAST. "f502n_DRZ.list" ick905030_drz.fits ick906030_drz.fits ick907030_drz.fits ick908030_drz.fits ick909030_drz.fits ick910030_drz.fits tweakreg.TweakReg('@f502n_DRZ.list',imagefindcfg={'threshold':10,'conv_width':3.5},refimage='f657n_sc04px10_drz_sci.fits', refimagefindcfg={'threshold':100,'conv_width':3.5},minobj=5, shiftfile=True,outshifts='shift502_drz.txt',sigma=2.5,searchrad=5.0,updatehdr=True,wcsname='DRZ',interactive=True,ylimit=0.5) from drizzlepac import tweakback tweakback.tweakback('ick905030_drz.fits',verbose=True) tweakback.tweakback('ick906030_drz.fits',verbose=True) tweakback.tweakback('ick907030_drz.fits',verbose=True) tweakback.tweakback('ick908030_drz.fits',verbose=True) tweakback.tweakback('ick909030_drz.fits',verbose=True) tweakback.tweakback('ick910030_drz.fits',verbose=True) Drizzle the set of 16 UVIS F502N *flt.fits frames using the F657N mosaic as a reference. astrodrizzle.AstroDrizzle('@f502n.list', output='f502n_sc04px10', clean=no, build=no, skymethod='match', driz_sep_bits='96', driz_cr_corr=yes, final_bits='96', final_wcs=yes, final_refimage='f657n_sc04px10_drz_sci.fits', num_cores=1, preserve=False) ----------------------------------- ----------------------------------- #Check products Blink the *drz_sci.fits and the *drz_wht.fits products for each filter to be sure there are no issues (eg. There should no signature of sources in the weight image, especially in regions where the tiles overlap). ds9 f110w_sc08px10_drz_sci.fits f110w_sc08px10_drz_wht.fits & ds9 f160w_sc08px10_drz_sci.fits f160w_sc08px10_drz_wht.fits & ds9 f502n_sc04px10_drz_sci.fits f502n_sc04px10_drz_wht.fits & ds9 f657n_sc04px10_drz_sci.fits f657n_sc04px10_drz_wht.fits & ds9 f673n_sc04px10_drz_sci.fits f673n_sc04px10_drz_wht.fits & Blink the drizzled mosaics to verify that the final products line up. This can be done by loading the 5 images and clicking 'zoom to fit' in DS9 for each one. This ensures that the frames are aligned in pixel space, albeit with a factor of 2 scale difference. Alternately, the user may click "align frames by WCS" to verify the accuracy of the updated image header WCS. ds9 f110w_sc08px10_drz_sci.fits f160w_sc08px10_drz_sci.fits f502n_sc04px10_drz_sci.fits f657n_sc04px10_drz_sci.fits f673n_sc04px10_drz_sci.fits & +--------+---------+---------+---------+---------+---------+---------+---------+ 4. Improving the UVIS calibration After aligning and combining the mosaics, we improved the UVIS calibrated data products (i*flt.fits) by correcting for CTE, crosstalk, and flat fielding artifacts as described below. These products were then redrizzled to create 'corrected' versions of the three UVIS narrowband mosaics. Note that the improved header WCS computed in Section 3 is preserved through this process and that the resulting mosaics will be aligned with the original set. A.) To correct the data for charge-transfer losses, we used the new Pixel-based Empirical CTE Correction Software for WFC3/UVIS data from the WFC3 page: http://www.stsci.edu/hst/wfc3/tools/cte_tools To perform the correction, both i*raw.fits and i*flt.fits products for visits 05-10 were placed in a separate directory. Next, the software was compiled and run as shown below. The data products are given a new naming convention: i*flc.fits. gfortran wfc3uv_ctereverse_parallel.F -o wfc3uv_ctereverse_parallel.e -fopenmp ./wfc3uv_ctereverse_parallel.e *raw.fits FLC+ Input: ick905k3q_raw.fits ick905k3q_flt.fits Output: ick905k3q_flc.fits B.) To correct UVIS data for crosstalk, a standalone IDL procedure is available to restore impacted pixels to a mean within ~1 sigma of the mean of surrounding pixels. The software and a description of its use may obtained from the WFC3 page below. (Click on the link for ISR2012-02 to download.) http://www.stsci.edu/hst/wfc3/ins_performance/anomalies Contents of the download directory: README WFC3-2012-02.pdf allframes_rgr_st.sav crosstalk_correct_wfc3.pro To run the code, start IDL and then correct the frames one-by-one, as shown below. The corrected images are written into the working directory as fits files with the same name as the original except the first character is changed to 'x'. idl crosstalk_correct_wfc3,'ick905k3q_flc.fits' crosstalk_correct_wfc3,'ick905k5q_flc.fits' .. .. Input: ick905k3q_flc.fits Output: xck905k3q_flc.fits C.) Rows flagged as bad in the flat field (DQ value=512) at the top and bottom edges of the 2 detector chips have been replaced with a value of 1000.0 for easier rejection by AstroDrizzle. Because these rows were populated with very tiny values (eg. +/-1e-10) in the flat, division by the flat field produces stripes with very high pixel and low values across the drizzled mosaics (+/-1e10 electrons) which are difficult to fully reject with cosmic-ray rejection. imreplace x*flc.fits[4][1:4096,1:4] value=1000.0 imreplace x*flc.fits[4][1:4096,2047:2051] value=1000.0 imreplace x*flc.fits[1][1:4096,2049:2051] value=1000.0 imreplace x*flc.fits[1][1:4096,1:3] value=1000.0 Finally, we redrizzle the full set of images per filter using the corrected data products. It's important to be sure the parameter 'updatewcs=False' when running AstroDrizzle or the improved WCS alignment derived in Section 3 will be wiped out. The three input lists are given below: "f502n_corr.list" "f657n_corr.list" "f673n_corr.list" xck905k3q_flc.fits xck905k5q_flc.fits xck905k8q_flc.fits xck905kaq_flc.fits xck905keq_flc.fits xck905khq_flc.fits xck905kjq_flc.fits xck905knq_flc.fits xck905kqq_flc.fits xck906kuq_flc.fits xck906kwq_flc.fits xck906kzq_flc.fits xck906l1q_flc.fits xck906l5q_flc.fits xck906l8q_flc.fits xck906laq_flc.fits xck906leq_flc.fits xck906lhq_flc.fits xck907niq_flc.fits xck907nkq_flc.fits xck907nnq_flc.fits xck907nxq_flc.fits xck907o1q_flc.fits xck907o4q_flc.fits xck907oqq_flc.fits xck907ouq_flc.fits xck907oxq_flc.fits xck908p9q_flc.fits xck908pbq_flc.fits xck908peq_flc.fits xck908pgq_flc.fits xck908pkq_flc.fits xck908pnq_flc.fits xck908ppq_flc.fits xck908ptq_flc.fits xck908pwq_flc.fits xck909c1q_flc.fits xck909c3q_flc.fits xck909c6q_flc.fits xck909cwq_flc.fits xck909d0q_flc.fits xck909d3q_flc.fits xck910d7q_flc.fits xck910d9q_flc.fits xck910dcq_flc.fits xck910deq_flc.fits xck910diq_flc.fits xck910dlq_flc.fits astrodrizzle.AstroDrizzle('@f657n_corr.list', output='f657n_sc04px10_corr', clean=no, build=no, skymethod='match', driz_sep_bits='96', combine_type='minmed',combine_nhigh=0, driz_cr_corr=yes, final_bits='96', final_wcs=yes, final_scale=0.04, final_pixfrac=1.0, final_rot=-35, final_ra=274.721587, final_dec=-13.841549, final_outnx=8000, final_outny=8400, num_cores=1, preserve=False) astrodrizzle.AstroDrizzle('@f673n_corr.list', output='f673n_sc04px10_corr', clean=no, build=no, skymethod='match', driz_sep_bits='96', combine_type='minmed',combine_nhigh=0, driz_cr_corr=yes, final_bits='96', final_wcs=yes, final_refimage='f657n_sc04px10_corr_drc_sci.fits', num_cores=1, preserve=False) astrodrizzle.AstroDrizzle('@f502n_corr.list', output='f502n_sc04px10_corr', clean=no, build=no, skymethod='match', driz_sep_bits='96', combine_type='minmed',combine_nhigh=0, driz_cr_corr=yes, final_bits='96', final_wcs=yes, final_refimage='f657n_sc04px10_corr_drc_sci.fits', num_cores=1, preserve=False) ----------------------------------- ----------------------------------- Copy to HLSP products using standard naming convention: cp f657n_sc04px10_corr_drc_sci.fits hlsp_heritage_hst_wfc3-uvis_m16_f657n_v1_drz.fits cp f502n_sc04px10_corr_drc_sci.fits hlsp_heritage_hst_wfc3-uvis_m16_f502n_v1_drz.fits cp f673n_sc04px10_corr_drc_sci.fits hlsp_heritage_hst_wfc3-uvis_m16_f673n_v1_drz.fits cp f110w_sc08px10_drz_sci.fits hlsp_heritage_hst_wfc3-ir_m16_f110w_v1_drz.fits cp f160w_sc08px10_drz_sci.fits hlsp_heritage_hst_wfc3-ir_m16_f160w_v1_drz.fits cp f657n_sc04px10_corr_drc_wht.fits hlsp_heritage_hst_wfc3-uvis_m16_f657n_v1_wht.fits cp f502n_sc04px10_corr_drc_wht.fits hlsp_heritage_hst_wfc3-uvis_m16_f502n_v1_wht.fits cp f673n_sc04px10_corr_drc_wht.fits hlsp_heritage_hst_wfc3-uvis_m16_f673n_v1_wht.fits cp f110w_sc08px10_drz_wht.fits hlsp_heritage_hst_wfc3-ir_m16_f110w_v1_wht.fits cp f160w_sc08px10_drz_wht.fits hlsp_heritage_hst_wfc3-ir_m16_f160w_v1_wht.fits +--------+---------+---------+---------+---------+---------+---------+---------+ 5. Data products Since these mosaics represent a significant investment of expert processing beyond the standard archival products, the AstroDrizzle-combined FITS images have now been released as a High-Level Science Product via the B. Mikulski Archive for Space Telescopes (MAST). We provide a list of downloadable FITS files which include both the drizzled science and weight FITS images, where the later provide an effective exposure time map for each filter. The filenames and descriptions of the drizzled mosaics are listed below. This file naming convention makes these HLSP files compatible and queryable with both MAST and the Hubble Legacy Archive (HLA). The format is: hlsp_project_mission_instrument_field-name_filter_version_product.extension hlsp_heritage_hst_wfc3-uvis_m16_f502n_v1_drz.fits SCIENCE IMAGE hlsp_heritage_hst_wfc3-uvis_m16_f657n_v1_drz.fits SCIENCE IMAGE hlsp_heritage_hst_wfc3-uvis_m16_f673n_v1_drz.fits SCIENCE IMAGE hlsp_heritage_hst_wfc3-uvis_m16_f502n_v1_wht.fits WEIGHT IMAGE hlsp_heritage_hst_wfc3-uvis_m16_f657n_v1_wht.fits WEIGHT IMAGE hlsp_heritage_hst_wfc3-uvis_m16_f673n_v1_wht.fits WEIGHT IMAGE hlsp_heritage_hst_wfc3-ir_m16_f110w_v1_drz.fits SCIENCE IMAGE hlsp_heritage_hst_wfc3-ir_m16_f160w_v1_drz.fits SCIENCE IMAGE hlsp_heritage_hst_wfc3-ir_m16_f110w_v1_wht.fits WEIGHT IMAGE hlsp_heritage_hst_wfc3-ir_m16_f160w_v1_wht.fits WEIGHT IMAGE +--------+---------+---------+---------+---------+---------+---------+---------+ 6. Citation The work described here is supported by the STScI Director's Discretionary Funds (Program 13926, PI: Levay) and the Hubble Heritage project. If you utilize these data products for scientific purposes, please reference: Levay Z.G., Christian, C.A., Mack, J., Frattare, L.M., Livio, M., Meyett, S., Mutchler, M.J., & Noll, K.S., 2015 AAS..22543602L, "New Hubble Space Telescope Multi-Wavelength Imaging of the Eagle Nebula" +--------+---------+---------+---------+---------+---------+---------+---------+