History of MISAO Project New Variable Star Discoveries
1999 Dec. 27
KenIchi Kadota, Seiichi Yoshida / MISAO Project
In 1997 April, Seiichi Yoshida, a graduate student of Waseda
University, made a plan to construct a database of enormous
astronomical images, set up a framework to discover new celestial
objects in the database, and develop a software to discover new
celestial objects as a theme of the master thesis. Because of the
diffusion of cooled CCD cameras, enormous images came to be yeilded in
the world, however, most of the images were not examined for new
objects and only little of data in the images were utlized. In order
to improve this situation, Yoshida started the MISAO (Multitudinous
Image-based Sky-survey and Accumulative Observations) Project, aiming
to make use of images in the world for new object discoveries and data
acquisition of known objects.
In Yoshida's thoughts, image examination must be automated in the
MISAO Project to deal with enormous images. So Yoshida started to
develop an automated image examination system which detects stars on
the images, compares them with data in CD-ROM star catalogs, measures
position and magnitude of all stars and picks up candidates of new
objects. This is the PIXY (PIXII: Practical Image eXamination and
Inner-objects Identification) system. Yoshida spent almost all of year
1997 for development of this software. Some people offered thier
images for experiments and Yoshida promoted the development with many
trials and errors based on the images. In this period, Yoshida did not
investigated the images for new objects because the system was still
poor. However, the system came to run roughly well in early 1998.
In 1998 March, KenIchi Kadota joined to the MISAO Project and started
a collaborated experimental survey and PIXY system improvement. In the
early stage, Yoshida and Kadota operated the survey at the
Hanadateyama Observatory (Bistar) in Miwa village, Ibaraki, using a
CCD camera under Kadota's instruction. However, the survey could be
run only in the weekend and the weather was not well, so the survey
succeeded only 4 times during 7 months between 1998 March and 1998
September. The about 200 images helped PIXY system improvement, but no
new objects were discovered.
After 1998 October, Yoshida and Kaodta purchased a CCD camera in
conjunction and Kadota started survey around his home in Ageo City,
Saitama, with his own instruments. Then the style is established that
Kadota takes CCD images and Yoshida examines them by the PIXY system.
At that time, hearing the discoveries of new variable stars by Kesao
Takamizawa or the discovery of the first new variable star by Katsumi
Haseda, survey for new variable stars came to be kept in mind.
In the early stage, 35-mm camera lens was used mainly for the survey.
After 1999 January, 180-mm camera lens was used to detect fainter
stars. But no new objects were discovered by the PIXY system comparing
two images of the same area with an interval of about one month,
although the ability for discovery of the system was already practical
and variable stars or ghost images looking like comets were certainly
detected. The framework was set up to create a database of magnitude
of detected stars by the PIXY system and compare them with data in new
images of the same area automatically. The check of known variable
stars was also automated.
In 1999 March, Yoshida examined the image of Sakurai's object (V4334
Sgr) taken by Kadota on Mar. 22, 1999 with 18-cm f/5.5 reflector
(990-mm focal length), by the PIXY system and discovered a 15-mag new
object. The object was not detected on the images of Sakurai's object
taken on Feb. 12, 1999 because fainter than 16.4 mag, so the system
detected it as a new object. This object was confirmed on Kadota's
follow up observation images taken on Mar. 31, 1999. It brightened to
14 mag and did not move. Then Yoshida announced it as the first new
variable star of the MISAO Project, MisV0001, on Apr. 3, 1999.
Discovery of new variable stars are authorized in public by being
published in the GCVS (General Catalog of Variable Stars) after
reporting to the Sternberg Astronomical Institute, Moscow University,
a consultative committee of the IAU (International Astronomical Union)
and the variability was confirmed. However, this style does not
function well recently because too many variable stars became to be
discovered to be managed by one organization. Therefore, variable star
discoverers generally named their own variable stars and publish them
in papers independently, and other discoverers and researchers
recognize them. In the MISAO Project, the new variable stars are named
as MisV (MISAO Variables) and published.
After that, some more new variable stars were discovered from the
18-cm reflector images for follow up observation of MisV0001.
At the same time, Kadota started survey with 16-cm f/3.3 reflector
(530-mm focal length) in 1999 April and some new variable stars were
also discovered from the survey images. Those new variable stars were
around 12-15 mag. The 16-cm reflector has 1.5 x 1 deg field of view
and can catch 16 mag star by 20 sec exposure. The PIXY system detects
about 2000 - 7000 stars automatically from one image in the Milky Way.
Considering these results, we realized that many new variable stars
would be discovered by taking continuous images in the Milky Way with
16-cm or 18-cm reflector. Then the variable star survey style is
established that taking images in the Milky Way with 16-cm reflector,
having wider field of view.
While 8 months between 1999 April and November, Kadota ran survey
30 times and obtained about 3000 images. The surveys were operated in
the open field and Kadota carried his instruments from home and
assembled them every time. The instruments were not controled by
computers, so all works such as taking sight, releasing a shutter,
saving image files, etc., are operated by hand. All survey images were
examined by Yoshida with the PIXY system. In addition, about 2000
images taken by Kadota with 18-cm reflector of comets and known
variable stars were also examined in the same way. The total number of
the stars detected by the PIXY system is about 15 million.
About 100 image are obtained by one survey and about 400 thousand
stars are detected. But in fact, there are two images of the same area
in order to reject noises, so the real number of stars is about 200
thousand. The PIXY system compares the detected stars with data
obtained from the past images of the same area and select only about
200 candidates of variable stars. Then Yoshida confirms these
selected candidates by eye on the images and removes uncertain or
blending ones. As a result, about 50 - 100 new variable stars are
For perfection, Kadota also checks the images of all new variable
stars discovered by Yoshida. This duplicated check style continues all
through our variable star survey and suspicious stars are not
announced. Because only stars with evident variability easily found by
eye on the images are announced, the magnitude range of most of the
new variable stars of the MISAO Project are larger than 0.7 mag.
Asteroids are also checked. Taichi Kato, Kyoto University, helps us on
identification with known variable stars. As a result of our
continuous survey, the number of new variable stars discovered by
Yoshida and Kadota in collaboration with the PIXY system reached to
739 at the end of 1999 December.