[Prev][Next][Index][Thread]

(meteorobs) Brightest lunar meteors probably large

     The objects that caused the brightest flashes that
we observed were probably a few hundred kilograms,
according to the messages below, since very little of the
impact energy is converted into light.  I've added a few
comments about the observations in the messages, using
" - " to preface my remarks.  
     There was a suggestion that sunglints from
artificial satellites might be involved, but this is
unlikely since the observations were made late at night
local time when most of these would be deep in the Earth's
shadow.  Also, with six events simultaneously recorded
at two or more separated locations, the chances are 
much greater that they are lunar phenomena than something
closer.  In the cases where lunar location information is
available in the separate video records, there is also
good agreement.
     Brian Cudnik reports that he observed from the
Houston Astronomical Society's site near Columbus, TX,
at long. 96 deg. 39' 50" W., lat. 29 deg. 37' 07" N.,
h 98m, about 100 km west of downtown Houston.
     There are many previous observations of probable
lunar impacts, although none of them apparently were
confirmed.  Many of these were published in a NASA
Technical Report on transient lunar phenomena that is
on the Web at http://www.mufor.org/tlp/lunar.html
One can also find there a link to a modern (about one
year old) effort to videorecord TLP's simultaneously
from different locations, a "lunascan" project, which
has other useful links, but so far they don't have
news of the Nov. 18th lunar impacts.  Apparently their
project has concentrated more on the terminator and
sunlit side of the Moon.  Also, published in the
Proceedings of the 48th convention of the Association
of Lunar and Planetary Observers (Las Cruces, NM,
June 25-29, 1997) is a good paper by John Westfall
on "Worthy of Resurrection: Two Past ALPO Lunar
Projects", including one on "Lunar Meteor Search"
that includes a table of meteor size, frequency,
flash magnitude, and crater diameter that is in
rather good agreement with the messages below, as
well as a good history of efforts up to 1997.
     Does anyone know of a Web (or other) reference to 
an account of the large impact observed by Canterbury
monks in 1178 that apparently caused the near-far-
side crater Bruno?  That's probably the first observation
of a lunar impact, although not confirmed from 
observations elsewhere.
     David Dunham, 1999 Dec. 7
_________________________________________________________

Date: Mon, 6 Dec 1999 11:30:36 -0700
To: Joan and David Dunham <dunham@erols.com>
From: Jay Melosh <jmelosh@LPL.Arizonadot edu>
Subject: References & calcs. of lunar meteor impacts
Cc: pweissman@issac.jpl.nasadot gov

Dear Joan and David:

I just heard from Paul Weissman that he estimates that your m = 3 flashes
must have been made by an object "about half a meter" in diameter.  I have
to agree with this estimate--which implies masses *much* larger than you
have mentioned!  (a half meter diameter projectile would mass about 500
kg).  The problem is that the luminous efficiency of an impact onto a solid
surface is *much* lower than the ca. 10% Mike Mazur estimates for a bolide.
This is discussed in detail in the Nemtchinov paper I mentioned in my last
email, but let me work out the consequences using Mazur's estimates for the
energy released by the various flashes on the moon (i.e.
L_obj=10^[(m-26.98)/-2.5] J/s, and an estimated duration of 33
milliseconds).  I use Nemtchinov et al.'s luminous efficiency estimate of
10^-4:

impactor mass,kg    crater diameter,m    luminous energy,J    Magnitude,m
      100                9.8                  2.5e7              4.8
      300               13.                   7.6e7              3.6
      500               15.                   1.3e8              3.0

I used my web program for computing crater sizes at
www.lpl.arizonadot edu/tekton/crater.html for the crater size computations,
assuming a projectile density of 1000 kg/m^3, impact angle of 45 degrees,
impact velocity of 71 km/sec and a target of loose sand (lunar regolith)
with a mean density of 2500 kg/m^3.

A potentially serious problem in these estimates is the duration of the
flash.  Nemtchinov and I computed that most of the light is emitted in a
single millisecond for a 1 m radius impactor--much shorter than the 1/30
sec Mazur estimates!  Smaller objects will produce correspondingly shorter
flashes.  However, I presume that your video camera (is it a CCD?)
   - Yes
integrates the light emitted over the duration of one frame (1/30 second?)
   - Actually, with interlaced video, the even lines are scanned in
     1/60th of a second, then the odd lines are scanned in the next
     1/60th of a second to form a 1/30th-second frame.  But some VCR's,
     including the ones we used, can work with the half-frames to
     achieve 1/60th second time resolution.
so Mazur's estimates for total energy emitted may be correct--but this has
to be verified before these estimates can be accepted.  If the actual
integration time was much smaller than assumed, that will reduce the mass
of the projectile fragment accordingly.
   - No, the integration time per half-frame is close to 1/60th second;
     as I understand, there is very little "dead time" between scans
     but there is some.  The E flash is curious in that I think it
     peaked between two scans in my tape, where it is almost equally
     bright, but rather faint, around 7th mag., on two successive
     half-frames.  But it must have been brighter, around 5th mag.,
     on a single half-frame for it to show up so well in Sada's tape.

The other uncertainty is that Nemtchinov et al. assumed an impact velocity
of 30 km/sec.  It is possible that the luminous efficiency is higher at 70
km/sec, and this is something that should be looked into, but it seems
unlikely it will be off by as much as a factor of 4.

I hope this is of help to you.
    - Yes, certainly, many thanks.

Sincerely,  Jay Melosh

###########################################################################

Jay Melosh                              Tel:   (520) 621-2806
Professor of Planetary Science          Fax:   (520) 621-4933
Lunar and Planetary Lab                 email: jmelosh@lpl.arizonadot edu
University of Arizona
Tucson AZ 85721-0092
_________________________________________________________

Date: Mon, 06 Dec 1999 20:00:35 -0800
From: R Clark <rclark99a@earthlinkdot net>
To: dunham@erols.com
Subject: size of lunar leonid impacts

Hello Dr. Dunham,

I was very excited to hear that several impacts on the lunar surface
had been detected during the Leonids. The possibility of such
observations has been examined several times over the years, and
generally ruled out as a difficult project with little likelihood of a
quick success. However the high efficiency and capabilities of modern
sensors and their widespread use by amateurs has now made the
observation a reality.

In the discussion of the observations you mention questions about the
size of the impactors that produced these flashes. You mention size
estimates ranging from >1000 kg to ~100 grams. I am curious about how
the latter figure reached you.
    - The lower estimates were from well-intentioned astrophysical
      calculations by others who, however, did not realize the
      very low fraction of energy that is transformed into visible
      light during these impacts.

For the size of the objects that produced these flashes, I have to
agree with the earlier figure... even though I am probably the source
of the latter. In my thesis at the University of Arizona I studied the
detectability of a different feature associated with lunar impacts.

High velocity lunar impacts produce several phenomena that may be
observed. The impact produces shockwaves in the target that may be
detected as seismic energy. The Apollo missions left a network of
seismometers which detected numerous impacts between 1969 and 1977 when
they, and the remaining active Apollo surface instruments were
foolishly shut down. (the old story about spending $40 billion to plant
a flag but not being able to afford the $50K/yr to receive and archive
the low but unending volume of science data still being returned)
Impacts of objects down to a few kg were detected with this networ
   - It sure would have been nice to have had ALSEP observations 
     of the Nov. 18th impacts!  Someone should have thought to try
     to look for flashes from separate observatories before shutting
     down the network.  Of course, the widespread availability of
     inexpensive sensitive CCD video cameras was key to this effort
     (the cameras we used only cost $80), and these didn't exist
     in 1977.

Another impact phenomenon, probably the most obvious thing to look for,
is the flash produced by the impact fireball. At velocities above ~12
km/sec (virtually all impacts of asteroidal or cometary material at
Earth) the impacting object and some ammount of target material
(increasing with higher velocities) will be vaporized to incandescent
temperatures. The radiation from this fireball will have its peak
intensity at visible or UV wavelengths, quickly dropping into the IR as
the gasses disperse and cool. The fraction of the impact energy
partitioned into the initial fireball is generally at most 10%. Only a
small fraction of this energy is released as 'visible' radiation while
the fireball gas is still hot and dense enough to radiate efficiently.
This has now been observed!

A very large fraction of the total impact energy (~60%) ends up as
thermal energy in the immediate vicinity of the newly formed impact
crater. This is what I was studying. After modeling cooling craters to
determine their radiative characteristics, I considered how to detect
them against the background of the cold lunar nightside with
groundbased, LEO, and lunar orbiting sensors. For space based sensors
the optimum wavelength range is in the 1-6 micron range. In the case of
a lunar orbiting sensor I concluded that an impact <100gm may be
detetable. For groundbased observations most of this wavelength range
is unavailable, although the 2 micron window might allow impacts of a
few kg or less to be detected, depending on scattering of light from
the sunlit portion of the disk and skyglow.

I am very pleased, and more than a little surprised, at how quickly the
(groundbased!) detection of any lunar impact events has come within the
grasp of modern instruments and sensors.

Richard Clark
rclark@lpl.arizonadot edu


Joan and David Dunham
7006 Megan Lane
Greenbelt, MD 20770
(301) 474-4722
dunham@erols.com
To UNSUBSCRIBE from the 'meteorobs' email list, use the Web form at:
http://www.tiacdot net/users/lewkaren/meteorobs/subscribe.html
Follow-Ups: