CCNet-ESSAY, 5 January 1999


By Brian G. Marsden <>

    One event that went essentially unnoticed during the millennial hype was
the reobservation on Dec. 13 of our old friend 1997 XF11. A pair of
positions of the object, of magnitude 20.6 at the time, was measured by the
original discoverer, Jim Scotti at Spacewatch, quite incidentally, while he
was observing the recent comet P/1999 V1, which was more than a magnitude
brighter and located about a quarter of a degree away.

    We know that the existence of the 1990 precovery observations by
Eleanor Helin and Ken Lawrence means that 1997 XF11 can be no conceivable
threat to the earth for several millennia to come. But what would the
situation be if nothing had been said about 1997 XF11 on 1998 March 11?
Almost certainly, there would have been no observations after those by Pete
Shelus at the McDonald Observatory on March 3 and 4. It is also quite
unlikely that anyone would yet have looked for the 1990 images--or for the
single, weak trail in 1995 that was found much later. Routine searching for
images of NEOs on old plates is a recent development brought about precisely
because of the 1997 XF11 "affair". Chances are that all we should have had
would be the 95 usable observations covering the 88-day arc following

    With that, it would have been possible to predict the position of
1997 XF11 in December 1999 to within an arcminute or two. The object
would surely have been identified--particularly if the 2028 close approach
had been recognized. In such a case, it is conceivable that the study of the
post-2028 motion I made during the months after discovery would in fact have
been carried out, with the appreciation that the 2028 approach could set
things up for an actual impact with the earth on 2040 October 26. Jim
Scotti's observations last month were just 3 and 5 arcseconds, respectively,
from this 2040-impact "prediction". The discordance between the two was
larger than normal because the image with the 5-arcsecond deviation was
affected by clouds. Not too surprisingly, the observations can be worked
into an impact solution anyway, with residuals of no more than 1 arcsecond.
The impact solution in 2040 requires a miss distance from the earth in 2028
of 0.0050 AU.

    Of course, the December 1999 observations were made on just a single
night, and it is standard policy not to announce second-opposition
observations without second-night confirmation. Presumably, if there had
been no 1990 data, that confirmation would have been quickly forthcoming. So
by the start of 2000 the observations would have been published, and new
orbit computations would have shown we are safe in 2040. Right?  Wrong! Even
with further observations in January, there would be insufficient
information to exclude the possibility of 2040 impact. On the other hand,
1997 XF11 might then have passed safely by at a distance of 0.3 AU. The
impact probability in 2040, 1 in 100 000 or better from the initial 88-day
arc, would be 1 in 10 000 or or better from the two-year arc. The event
would have rated a clear 2, maybe even approaching a 6, on the Torino
Scale--or as I prefer to state it, in terms of the colors, rather than the
numbers with their undue sense of precision, we should have had a light
yellow, perhaps a light orangey-yellow, alert.  As stated, the precovery
data completely remove the threat, and the miss distances in 2028 and 2040
are 0.0064 and 0.18 AU, respectively.

    Together with the search for precovery images, the routine computation,
notably by Andrea Milani and Steve Chesley, of impact possibilities for
known objects during the next half-century or so is another beneficial
outcome of the experiences with 1997 XF11. Since my main concern, in
initially drawing attention to the 1997 XF11 situation, had been the general
apathy that greeted the discovery of an obviously interesting NEO, I am much
encouraged that they took up the recommendation I made at an IAU symposium
in July 1998. Independently of this, the "impact" of LINEAR on the NEO scene
in April 1998 was another extremely welcome development, augmenting the
discovery rate by at least a factor of five. Immediate and enthusiastic
follow-up observations, by a handful of professional groups and an
ever-increasing number of amateur
astronomers around the world, have eagerly kept up with this activity.
Later follow-up and second-opposition recovery have been more
troublesome, but professional colleagues such as Dave Balam, Rob McNaught,
Carl Hergenrother and Jana and Milos Tichy continue to play an important
role with these. In terms of NASA's aim of 90-percent completeness in the
"detection" (if not the follow-up) of kilometer-sized NEOs, recent work by
Dave Rabinowitz suggests that past overestimates of the number of objects
will allow the job to be done in 15 years or so, not outrageously longer
than the 10-year specification.

    This is a time to look ahead to the next phase in NEO discovery, when
more attention will necessarily be given to objects smaller than 1 km
across. Spacewatch has already started to pave the way, but the sky coverage
for faint objects needs to be greatly extended. Longer exposures by LINEAR
can help, but at some point larger instruments will be needed, and in both
hemispheres, as discussed in the original Spaceguard Survey document in
1992. The problem is more severe for follow-up and recovery, already a
difficulty for Spacewatch discoveries beyond the reach of amateurs.
Although, as now, a lot of the follow-up can be automatically achieved if
imaging of the fields is repeated on several nights each month, the fainter
recoveries require ready access to large telescopes (and precoveries are out
of the question for faint objects). In this connection, it is indeed
gratifying to see that some countries, notably the U.K., are considering the
establishment of national Spaceguard centers. If different nations work
together in an international framework, there is obviously a greater
likelihood that all aspects of the "sub-km" phase can be satisfactorily

    In conclusion, I should say that I think the NEO community as a whole
still has a lot to learn about communicating information on possible NEO
threats to the public. We have all learned a lot since the 1997 XF11
affair, and the Torino Scale, while perhaps a start, needs some
modification, particularly so that it includes a parameter showing the time
elapsing until the possible impact event. The Pisa "risk page" is useful,
but it is basically for informed readers: nevertheless, it is obviously also
going to be seen by people who don't understand the difference between an
observation and a hypothetical calculation. They will surely have questions
the Torino Scale does not answer. It was also unfortunate that the 1999 RM45
risk page remained unchanged for some 48 hours after new observations and
the resulting orbit computations (the computations nicely documented on
associated pages) clearly removed the suggested 2042 threat. Yet, after
that, the risk page was completely removed, just as the 1999 AN10 risk page
had been removed in July, once it could be ascertained that the object could
not be a threat during the 2040s and the following few decades. This is not
right either, because it removes all traces of the past history. We learn
from history, and to pretend that it does not exist really does not help us
to move forward and allow us to get the NEO problem under control.

CCCMENU CCC for 2000