Catching the Shadow of Kuiper Belt Object The TAOS Project by nyut545e2


									                                                                    Catching the Shadow of Kuiper Belt Object: The TAOS Project

Catching the Shadow of Kuiper Belt Object:
The TAOS Project*
Typhoon Lee
Academia Sinica Institute of Astronomy and Astrophysics, Taipei, Taiwan

                                                 has been arguably the most active topic in          ASIAA, Inst. Astronomy of National
                                                 planetary science during the past two de-           Central University (both in Taiwan),
                                                 cades. This period coincides with Taiwan’s          Harvard-Smithsonian Center for As-
                                                 rapid growth in astronomical research and           trophysics (in the U.S.) the Astronomy
                                                 her breakout of the national boundary in            Dept. of Yonsei University in Korea and
                                                 search of powerful instruments at superior          several individuals from other institutes
                                                 sites with the most favorable weather               the survey aims to detect the occultation
                                                 condition. Its centerpiece for
                                                 KBO research was the TAOS
                                                 project which stands for: “Tai-
                 Typhoon Lee                     wanese-American Occultation
                                                 Survey” (for KBO). TAOS has
Our knowledge of the solar system beyond         demonstrated that by using the
Neptune used to consist of a tiny planet,        most up-to-date computer-con-
Pluto, and an inferred but not detectable        trol-communication technol-
spherical shell of comets, the Oort cloud.       ogy, a group of small robotic
Large telescopes, large CCDs, and large          telescopes deployed on the 3
comets changed all that. Over one thou-          km high Lu-Lin peak in central
sand objects, all larger than the largest        Taiwan can carry out sophisti-
comets (< 30 km) seen near earth, have           cated automatic observations
been found so far between 30 and 80              to accomplish internationally
AU. Instead of an oddball planet, Pluto          competitive research.
became (almost) the largest member of
this population which is now known as      TAOS is an international Fig. 1: The diameter vs. distance plot of KBOs. The
                                                                    reflected sunlight received on earth decreases as the
the Kuiper Belt Objects (KBO). KBO collaboration consisting of 4th power of the distance because in each leg of the
                                                                                          travel, light obeys the inverse square law. Such a fast
                                                                                          drop makes it hard to observe even the largest KBOs (>
 Further information on the TAOS project can be found in the following publications       1,000 km) much beyond 100 AU since the most powerful
and at its homepage                                     telescopes on earth have a limiting red band magnitude
                                                                                          less than 30 (the “R=30” line in Fig. 1). For the same
The first result is in:                                                                   reason it is also difficult to observe KBOs whose sizes
• First Results From The Taiwanese-American Occultation Survey (TAOS)                     are close to that for near earth comets (around 10 km)
(by Z.-W. Zhang, F. B. Bianco, M. J. Lehner, N. K. Coehlo, J.-H. Wang, S. Mondal,         even at the inner edge of the Kuiper Belt near Neptune
C. Alcock, T. Axelrod, Y.-I. Byun, W.-P. Chen, K. H. Cook, R. Dave, I. de Pater, R.       (about 35 AU). The R=30 is the dividing line beyond
Porrata, D.-W. Kim, S.-K. King, T. Lee, H.-C. Lin, J. J. Lissauer, S. L. Marshall, P.     which lies the large niche where only the occultation can
Protopapas, J. A. Rice, M. E. Schwamb, S.-Y. Wang, C.-Y. Wen)                             reach. The other limit for occultation is set by the finite
Astrophysical Journal 685, L157 (2008)                                                    size of the background star and the diffraction fringes
(DOI: 10.1086/592741)                                                                     around the edge of the KBO. Diffraction pattern around
(preprint is available as arXiv:0808.2051v1 [astro-ph])                                   an aperture becomes most prominent when the size of
                                                                                          the aperture becomes comparable to the geometrical
The description of the system is in:                                                      mean of the wavelength and the distance to the KBO.
• The Taiwanese-American Occultation Survey: The Multi-Telescope Robotic                  Since the diffraction effect dominates before 2,000 AU
  Observatory                                                                             and only weakly dependent on the distance (i.e. recall
(by M. J. Lehner, C.-Y. Wen, J.-H. Wang, S. L. Marshall, M. E. Schwamb, Z.-W.             that the exponent is 0.5 instead of 4). Therefore >1,000
Zhang, F. B. Bianco, J. Giammarco, R. Porrata, C. Alcock, T. Axelrod, Y.-I. Byun,         km sized KBO can be detected at 1,000 AU although the
W. P. Chen, K. H. Cook, R. Dave, S.-K. King, T. Lee, H.-C. Lin, S.-Y. Wang)               event rate will be extremely low. However the occultation
Publications of the Astronomical Society of the Pacific 121, issue 876, 138 (2009).       technique has one important shortcoming, namely the
(DOI: 10.1086/597516)                                                                     KBO found this way cannot be tracked thus it is impos-
(preprint is available as arXiv:0802.0303v1 [astro-ph])                                   sible to determine its orbit.

                                                                                        AAPPS Bulletin October 2009, Vol. 19, No. 5 21
Astronomy in Taiwan

         Fig. 2: (a) An occultation event occurs when a KBO passes between the telescope and a distant star. (Picture credit LLNL).
         (b) Shadow of a KBO projected onto the surface of the Earth. Note the significant diffraction effects. The image is 10 km
         on a side.

of distant background stars by KBOs with        night with all four telescopes to search        that we make as many as several billion
size comparable to near earth comets (i.e.      for coincidental flux variations consistent     photometric measurements per year on
a few km in diameter). The idea is de-          with occultations by KBOs. TAOS is also         a single telescope, special care must be
ceptively simple as depicted in Fig. 2(a).      sensitive to more distant objects out to        taken to minimize the rate of false positive
When a KBO occasionally moves across            1,000 AU since its dependence on the            events. We thus require coincident detec-
our line of sight to a distant star, its        distance is only 0.5 (i.e. the square root),    tion of any events on all four telescopes.
shadow blocks the star light, just like the     which is much smaller than the exponent         In this way we keep our false positive rate
moon passing in front of the sun causing        of 4.0 for reflected light. The discovery of    below 0.1 events per year.
an eclipse. Since a KBO typically has a         Sedna at 80 AU implies the existence of
diameter of only a few km (say 6) while         a hitherto unknown population of objects          We began the survey using only three
the shadow speed is close to 30 km/sec on       beyond 100 AU, which are too distant            telescopes in February of 2005, and
the ground mostly due to earth’s orbital        to be perturbed by any known planets            four-telescope operations commenced
motion, the star’s blink only lasts 6/30 ~      at their present positions. The estimated       in August of 2008. To date we have col-
0.2 sec. i.e. we have to sample the light       KBO occultation rate is extremely low           lected over 10 billion three-telescope
curve at 5 Hz. Such a short integration         and highly uncertain. Predicted rates range     photometric measurements and more than
time means high noise. Note also that the       from 0.01 to 100 events per year. Given         3 billion multi-telescope measurements
typical size of a few km for KBOs happens
to be close to the geometrical mean of our
distance to the KBOs and the observing
wavelength (about 0.6 μm). In this so
called Fresnel regime, the strong diffrac-
tion pattern consisting of interference rings
[Fig. 2(b)], will further increase the noise
when the sampling rate is lower than 4-5
Hz. However, at high sampling rate (15-30
Hz) one may hope to confirm that a dip in
the light curve is due to KBO occultation
and learn more about KBOs.

  TAOS operates four fast and wide-field
0.5 m f/2 robotic telescopes at the 3 km
high Lu-Lin Observatory in central Taiwan
[Figs. 3, 4]. Each telescope is equipped
with a 2k×2k CCD camera which is read
                                                       Fig. 3: One of the TAOS 0.5m f/2 telescopes in its enclosure. Its unusually small
out with a cadence of 5 Hz. We thus moni-              f ratio requires extremely accurate finishing of the primary mirror and a compli-
tor 500 to 1,000 stars simultaneously per              cated lens to correct aberration plus very tight tolerance on the alignment.

22   AAPPS Bulletin October 2009, Vol. 19, No. 5
                                                                Catching the Shadow of Kuiper Belt Object: The TAOS Project

with all four telescopes. The system oper-    we expect to collect, TAOS II should be          cultation shadow, which will allow us to
ates in fully automatic mode, with remote     capable of probing the entire parameter          make estimates of the sizes and distances
monitoring from ASIAA and NCU. A              space predicted by models of the size            of any detected objects. We expect TAOS
dress rehearsal of the system was carried     distribution. Furthermore, given the higher      II to start collecting data in 2012, and we
out in June 2004 when an 8.5 mag. star        readout speed, TAOS II will be able to           plan to operate the survey for a total of
was occulted by a 15.5 mag. asteroid          resolve the diffraction fringes in the oc-       four years.
(#1723 Klemola, diameter 31 km). Two
TAOS telescopes successfully detected
this event with better than 0.25 second
time resolution under remote control
[Fig. 5]. The team has analyzed all of the
three-telescope data through December
of 2008. No candidate events were found,
and TAOS has placed the strongest upper
limits on the size distribution of objects
with 0.5 km < D < 28 km that have been
published to date. (The TAOS limits are an
improvement of three orders of magnitude
over the most recently published limits.)
Furthermore, TAOS is exploring several
different models for the formation of the
Kuiper Belt. Given that no events were
found, we can exclude at the 95% confi-
                                                       Fig. 4: The four TAOS telescopes on top of Lu-Lin Mountain. Can you find
dence level any theoretical model which                them? A and B are in the foreground, C near the center of the picture, D
predicts that TAOS would have detected                 close to the flagpole (far left). Telescope separation: 6 -60 m.
three or more events. This is illustrated
in Fig. 6, where some example models
of the size distribution are shown. These
models predict anywhere from a few to
a few thousand events to be detected by
TAOS. This shows that TAOS is capable of
                                               Fig. 5: TAOS images (two telescopes) of the occultation of HIP050525 (mv~8.46 mag) by
probing a significant amount of parameter      the asteroid (1723) Klemola (mv~15.7 mag; D~31 km). Each frame is a 0.25 second read out
space, and continued operation will allow      of the field (time increases to the right.) Some 10 billion stellar photometric measurements
us to place even more stringent constraints    have now been made in this shutter-less “zipper” mode by TAOS.
on these models in the future.

   We have also begun design work on
TAOS II, a successor survey to TAOS. The
design goal of TAOS II is a factor of 100
improvement in the event rate over TAOS.
This will be achieved by using larger
telescopes, a higher imaging cadence (20
Hz), and a better site (Mauna Loa, Hawaii;
Baja, California; Mt. Hopkins, Arizona; or
Las Campanas, Chile). The current plan
is to install three 1.3 m F/4 telescopes on
Mauna Loa in Hawaii, and equip each
telescope with a custom camera, each
consisting of an array of fast readout
EMCCD chips. TAOS II will collect over         Fig. 6: Left panel: A series of models of the size distribution of the Kuiper Belt Objects pub-
a petabyte (1,000 terabytes) of raw image      lished by Kenyon and Bromley (2004, AJ 128, 1916). Points represent measurements of the
                                               size distribution from direct observations. Right panel: The cumulative number of expected
data every year! Given the improvements        events from the TAOS survey for the models in the left panel. Any model which predicts more
over TAOS and the vast quantity of data        than three events (dashed line) is excluded at the 95% confidence level.

                                                                                 AAPPS Bulletin October 2009, Vol. 19, No. 5 23

To top