XRT synoptic images : Improvement and Archive

1. Improvement of image quality

1-1. From 2-image to 3-image composites.

Source data (synoptic data set at 2013/06/07 06:07), click to enlarge.

short exposure medium exposure long exposure

Open/Ti-poly 0.0327 sec Open/Ti-poly 0.2560 sec Open/Ti-poly 1.0255 sec

Maps of saturated pixels (click to enlarge).

short exposure medium exposure long exposure

Resulting composite images (click to enlarge).

short + long medium + long short + medium + long

Close-ups of the Active Region.

[ memo ] Images prepared with TRIPLE_SYN_DEMO.PRO.

1-2. Correction of the contamination spots.

XRT_SPOTCOR.PRO (provided by S.Sarr at SAO, 2-Sep-2013) applied to the composite synoptics images (click to enlarge).

sample 1 : Al-mesh image (left: before application, right: after)

sample 2 : Ti-poly image (left: before application, right: after)

1-3. Treatment of missing frames.

Modification of MK_XRT_COMP3.PRO to handle missing frames, i.e.,

missing frames in Long image : replaced with corresponding frames in Medium image.
missing frames in Medium image : replaced with those of Short image.
missing frames in Short image : replaced with those of Medium image.

Modification of MK_XRT_COMPOSITE.PRO.

missing frames in Long image : replaced with corresponding frames in Short image.

old MK_XRT_CMP3 new MK_XRT_CMP3

1-4. Correction of pointing info in the FITS header.

The following FITS tags about the pointing in the official XRT level0 products are not precise enough to give exact location of the limb or sun center coordinates.

Sample of pointing tags in a Level_0 FITS file (XRT20130718_181542.5.fits):

hdr_l0.CRVAL1          DOUBLE          -40.000000
hdr_l0.CRVAL2          DOUBLE          -22.145500
hdr_l0.CROTA1          DOUBLE          0.69997900
hdr_l0.CROTA2          DOUBLE          0.69997900
hdr_l0.XCEN            DOUBLE          -40.000000
hdr_l0.YCEN            DOUBLE          -22.145500

Fortunately, Keiji Yoshimura succeeded in correcting the XRT pointing info by applying limb fitting to the synoptic images. I used his results (and his procedure, KY_XRT_READ_COALDB, to access to his database) to correct the above FITS tags for the FITS files of the synopcic composite images.

IDL> coaldb_dir='/disk/users/yosimura/XRT_Coal/Limbfit/coaldb.seq02'
IDL> hdr_scia=ky_xrt_read_coaldb(hdr_l0,top=coaldb_dir)

Corrected pointing tags in the SCIA FITS header for the corresponding data (comp_XRT20130718_181542.5.fits):

hdr_scia.CRVAL1          DOUBLE          -25.872089
hdr_scia.CRVAL2          DOUBLE           18.439325
hdr_scia.CROTA1          DOUBLE         -0.29034933
hdr_scia.CROTA2          DOUBLE         -0.29034933
hdr_scia.XCEN            DOUBLE          -25.872089
hdr_scia.YCEN            DOUBLE           18.439325

Plot_map of Level_0 (left) and SCIA (right) FITS files.

2. Synoptic Composite Images Archive.

current synoptic image archive at SAO
New synoptic image archive under construction at MSU (my project!)
- design the specification.
- determine the data directory sructure.
```
     mithra:/disk/yla/takeda/xrt_synoptics> ls
     demo_PNG/   syncmp_FITS/  syncmp_browz/  syncmp_thumb/
     syn_files/  syncmp_PNG/   syncmp_gen/    thumb_work/
     
```
- prepare a proto type, collect comments from the team, & determine the final shape.
- prepare a code for auto generation, for where possible.
Here is the new Synoptic Composite Image Archive (SCIA).
- new feature:
  - calender-like monthly pages with a link to the daily pages.
  - a navigation bar on left side.
  - "operation notes" on each monthly page, explaining special operations, anomaly, reason of missing images, etc.
- troubles encountered :
  - glitches in XOB and FITS header, e.g.
    - long/short exposures have different resolution,
    - irregular order of exposures, short/long/medium,
    - some FITS header lack the PIXCNT tag, used for handling missing frames.
    - descrepancy between FITS tags, EC_FW1 and EC_FW1_ (in Dec-2006).
    - erroneous value of P1COL, P2COL tags (Dec-2006).
  - difficulty in eliminating unwanted images (off-points, etc).
  - difficulty in tracking the reason of missing data from the CO e-mail archive.
    - only partial info found in the CO archive (time of emergency, etc.)
    - no info 2006--2007/03 (as CO_archive starts Apr-2007).
  - too many 2048x2048 images in 2006-2007, result in memory shortage on Ylstone.
- specal thanks to :
  - Steve Saar for providing XRT_SPOTCOR.PRO. It sure improved the image quality.
  - Jon Slavin for preparing improved synoptic FITS files. I am looking forward to download them to add to the SCIA.
  - Sakao-san and C-soda folks, for upgrading the XRT_CO mail archive. The new e-mail archive really helped to track the special operations in the past.
  - a scientst at SAO who reviewed the initial version of SCIA and reported errors and missing links.
  - Keiji Yoshimura for providing KY_XRT_READ_COALDB.PRO. It improved FITS header info about XRT pointings.

short exposure	medium exposure	long exposure

Open/Ti-poly 0.0327 sec	Open/Ti-poly 0.2560 sec	Open/Ti-poly 1.0255 sec