First results of the occultation of 32 Lyncis by (372) Palma, Friday Morning, Jan. 26, 2007

At least 20 observers timed the occultation in spite of the actual path shifting nearly half a path-width south of the "Jan. 12 baseline" predicted path

Although it was the coldest night of the year in many areas, it was clear over most of the path across the USA; only a few observers were clouded out

Updated: 2007 February 2 UT, 2h UT (Feb. 9 pm EST)

The actual path was about a third of a path-width south of the last 
predicted path (Steve Preston's update of Jan. 25), which in turn 
was 26 km south of Preston's Jan. 12th update that was used as the 
"baseline" for planning coverage of the event.  Also, since Palma 
was elongated approximately in the direction of motion, the path was 
a little narrower than predicted.  Although this caused about half 
of the mobile stations to be too far north and have no occultation, 
there was very good coverage of the central and northern part of 
Palma, and a few observers who tried to observe at fixed sites south 
of the predicted path were rewarded with an occultation and obtained 
some valuable timings on the southern part of Palma.  The timings of 
the 20 observers who timed the event, as well as the paths for 
several observers who were too far north to have an occultation (but 
did show that no satellites of Palma larger than about 4 km could 
have been on the northern side of the asteroid), are shown in 
this plot, also included in this Word file for possibly easier 
viewing and printing for some.  The plot shows the plane of the sky 
at Palma; the lines show when the star was visible for the 
observers.  The disappearances are on the right side of the plot.  
The dashed line marked "21" is the line for a theoretical "observer" 
at a point in Steve Preston's predicted central line as last updated 
on Jan. 25.  Some of the timings are preliminary and will be refined 
during the next several days.  A few of the preliminary visual 
timings are discordant relative to the other observtions and so were 
given zero weight in the solution for the fitted ellipse, whose 
major and minor axes were found to be 207 and 185 km +/-2 km.  
Observer 35 only noted that the occultation lasted "1 to 3 seconds" 
at his location; that is remarkable considering the accurately timed 
duration of 6 seconds measured from a drift-scan CCD image at the 
southernmost station #36.  Either the asteroid has a strange shape, 
or more likely the occultation was actually longer than what 
Observer 35 thought.  The observations are in this plain text file 
and in this WinOccult ".OBS" file.  The plot was prepared by Bradley 
Timerson and myself.  Next week, a more up-to-date plot, including 
any new reports received, can be found on Brad Timerson's North 
American asteroidal occultations results Web page; it will also 
include maps showing the observer locations and other details.

THE ACTUAL AND PREDICTED PATHS ACROSS THE USA

The actual and predicted paths of the occultation across the USA are 
shown on the first slide of this Power Point file.  The green line 
is the last predicted central line by Steve Preston, computed the 
day before the occultation (Jan. 25), while the blue lines are the 
predicted northern and southern limits.  The two gray lines show, 
within an accuracy of about +/-2 km, the northern and southern 
limits of the actual path determined from the observations.  Since 
the actual path was south of the predicted path by a significant 
amount, locations where no occultation was expected actually had an 
occultation, including Roanoke, VA; Lexington and Louisville, KY; 
St. Joseph, MO; Greeley and Longmont, CO (and the northernmost part 
of Boulder); Provo, UT; Reno, NV; and Grass Valley, Marysville, and 
Ukiah, CA.  In the Power Point file, slides two to five show the 
path in more detail relative to major highways and cities on 4 
separate maps roughly corresponding to the path across the four 
continental USA time zones.  All of these maps were created from 
Brad Timerson's interactive google maps Web page that you can use to 
see where the limits were in much more detail.  In case you can't 
open or download the Power Point file, the map image files are 
listed below.

Paths across USA,                      
Eastern Time Zone,  Delmarva to Indiana 
Central Time Zone,  Illinois to Nebraska
Mountain Time Zone, Colorado to Utah
Pacific Time Zone,  Nevada to California

The size of the southward shift of the actual path from the 
predicted one was larger than we usually have during asteroidal 
occultations, about 2.5 standard deviations in this case, which 
means there was a little less than 1% chance that such a shift would 
occur.  Probably there was a systematic error that was not taken 
into account in the statistical analysis.  Two other occultations by 
(372) Palma were observed well enough to determine the paths, in 
2000 and 2005, both in Europe; those show that the orbit of Palma 
that Steve Preston used for the prediction had an error of 0.015" or 
less, while the shift we had was over 0.050".  So it seems likely 
that there is an obscure problem with the proper motion of 32 Lyncis 
as determined by Hipparcos.  A few other large errors like this in 
the past have been found to be due to duplicity of the star, but the 
video occultation observations showed no sign of duplicity of 32 
Lyncis.  Problems have been found with the Hipparcos positions of 
Pleiades stars, apparently due to the weighting scheme used that 
assumed a uniform distribution of bright stars, not the clustering 
that occurs with the Pleiades.  Could nearby 33 Lyncis have caused a 
problem the same way with the positions of both it and 32 Lyncis?  
We are asking the astrometry experts who might know about these 
possible problems so that we might assign more realistic "error 
bars" for future occultations of bright stars.
_______________________________

I described my effort to observe the Jan. 26th occultation of 32 
Lyncis by (372) Palma from 4 widely-separated stations in Virginia 
in the message below, using pre-pointed telescopes at 3 locations 
and observing the usual way at the 4th one; all observations were 
video.  The stations were at or near Cold Harbor (just east of 
Richmond), Bowling Green, Port Royal, and Dahlgren.  The actual 
locations were within a km of the planned ones in terms of distance 
from the predicted central line. I have completed review of the 
tapes and have sent the details to reports@asteroidoccultation.com , 
to which all North American asteroidal occultation observation 
reports should be sent, preferably using the forms that you can 
obtain here.  Also on that Web site are posted the observational 
results, including sky plane reconstructions of the shape of the 
asteroid ("Occult"), maps showing the predicted path and observer 
locations, now color-coded to indicate positive and negative 
observations, and the observational data.                        

As expected from the other reports received, which showed that there 
was a large south shift from the prediction, I obtained one positive 
(occultation) and 3 negative (miss) observations.  At least 
everything worked in the field in spite of the long record times and 
19 deg. F. temp., the first time I've successfully recorded the 
target star at the right time from 4 widely separated locations.  I 
just wish that more of them were positive, and that I didn't hit the 
"record" tab by mistake when trying to get the accurate times for my 
positive observation at Cold Harbor, one of my nicer asteroidal 
occultation recordings gone.  At least, I did get useful preliminary 
less accurate times from the tape before that disaster.  The reports 
for my stations, all in Virginia, from south to north (so the 
positive result is the first one listed), in the now standard Excel 
files, are below: 

Cold Harbor, remote 1
Bowling Green, remote 2
Port Royal, remote 3
Dahlgren, attended

Although I lost my recording of the occultation, several other 
observers have theirs.  Derek Breit has posted his video recording 
of the event at the bottom of his Web page devoted to results of the 
Palma occultation.

While everything worked quite well for me using old equipment that 
morning, I was surprised and saddened that several observers, 
including some rather experienced ones, failed to make an 
observation for one reason or another.  This was a bright star, 
quite rare for an asteroidal occultation, and those who failed 
should have had a backup plan when things started to go wrong with 
their video equipment or telescope.  Paul Maley always takes a pair 
of binoculars with him when he travels for events bright enough to 
see with them, and that has saved his observation more than once.  
When I arrived at my last location, not the intended site but one 
selected hastily (a long driveway to a church) close to the major 
highway because it was less than 25 minutes to the event, I was 
afraid that I would have to just use a finder scope and observe 
visually. But I managed to get the video working and started 
recording 32 Lyncis there with a 4" Meade SCT 3 min. before the 
closest approach time.  For those who didn't observe entirely, or 
partly, because of the cold, sorry, they're wimps. But even they 
could have made a useful observation from home, where they could set 
up their telescope early and warm up, staying outside for relatively 
short periods, or they could just observe the event with binoculars 
- the star was bright enough that they could have done that looking 
out a window from inside their house, or better, from their car 
(which would allow mobility, to possibly get closer to the predicted 
path, or at least to a relatively dark location).  Any observation 
is better than no observation. 
___________________

My earlier message is below:

From: David Dunham [dunham@starpower.net]
Sent: Tuesday, January 30, 2007 11:15 AM
To: Dunham, David
Subject: 4 stations deployed across VA; positive at Cold Harbor

- For the first time, I ran by myself 3 remote stations, plus an 
attended station, all at locations at least 12 km apart [actually, 
they were 18 km or more apart] from each other; the total spread was 
about 87 km across the path, from -19 to +68 km.  I've set up 4 
stations at least a couple of other times in the past, but didn't 
get them all to work, except for the Alexandra event in Baja Calif., 
where I had help from local observers, including some who stayed to 
run two of the "remote" (pre-pointed) telescopes (that event also 
had a large south shift, and all 4 of my stations for it had a 
miss, but one of them was clouded out a few minutes before that 
event). The tapes for Palma at the middle two stations, at Bowling 
Green and north of Port Royal, did record well after the predicted 
time, but I have not had time to review those tapes; they must be 
misses since they were well north of Paul Maley's observations near 
the central line, but maybe they caught a satelloid - I'll check 
them after getting some sleep, none of which I got last night (I did 
sleep from 11 am to 2 pm after returning home).  Altogether, I drove 
505 miles since I had to cover most of the spread 6 times (3 round 
trips). 

     I thank Ed at Ed's Landscape Center in Cold Water for letting 
me observe from his nice large dark spread a third time, this time 
with success.  The tape there had to record for two hours, so I 
started it at 7:20 UT and then drove north to attach and turn on 
recorders and time inserters at the other two telescopes that were 
set up and pre- pointed earlier. But in fact the Sony Video Walkman 
recorded for nearly four hours; Ed loaned me an extension cord (I 
meant to bring one, but forgot to at the last moment as I left home) 
so I didn't need to depend on a battery for it.  Anyway, it was 19 
km south of the central line of the baseline Jan. 12th prediction.  
32 Lyncis entered the video FOV at 9:43:19 UT and exited it at 
9:44:45.  372 Palma occulted the star from 9:44:01 to 9:44:12 (Palma 
was faintly visible while the star was occulted); I'll get precise 
times from the Kiwi time insertion later, and send the report to 
Brad Timerson.  A 20cm SCT with f/3.3 focal reducing lens and PC164C 
camera was used. 

     By the time I started all the remote stations, I didn't arrive 
in the Dahlgren area until about 22 min. before the occultation, and 
in fact didn't observe from my intended location a mile from US 301 
but instead just pulled into a long church driveway very close to US 
301 but a couple of hundred feet from traffic, and the nearest 
street lights.  Beforehand I figured my best success there would be 
with a 4-inch SCT with my image intensifier, since that would give a 
1-deg. FOV, helping in locating the target star.  While at the 
southern station waiting for a pre-point opportunity, I had time to 
screw the 3 short legs into the 4-inch scope's base and attach the 9 
x 50 finder scope.  My minivan doesn't have a suitable surface on 
which the scope could be set; the roof is too high and other parts 
too curved.  I tried setting it on a 5-inch scope case, but it was 
too small and the scope fell off it; I ended up just putting it on 
the ground and getting low enough to see through the finder. I used 
a digital camcorder that I had been using earlier for setting up the 
remote stations, so it had recordings of GPS time insertion that 
would calibrate the digital camcorder's clock, so I didn't need to 
take the time to set up a time inserter there, although I did have 
another one.  I did have to align the finder scope with the main 
scope, with the image intensifier.  At first, I tried a random star, 
but that didn't work; the one in the center of the finder field 
wasn't the one in the video field.  Stars of 2nd mag. and brighter 
are too bright for the image intensifier (could damage it), so I 
couldn't use Saturn or Regulus.  So I moved to the Praesepe, an 
easily-recognized large group of stars, and moving around just a 
little found it with the video; then I could align the finder well 
enough.  By then, there were only a few min. to go, I thought not 
even enough time to consult the star charts. But I remember from 
working with the charts about what to expect working from alpha 
Lyncis and the 4th-mag. stars near it, which I could see naked eye 
from there.  So I quickly star-hopped from them, finding the line of 
stars east of the 32 - 33 pair, and then the distinctive pair 
itself.  I turned on the recorder, but neither star in the pair 
dimmed.  A couple of minutes after the closest approach time, I 
brought the charts over to the camera and verified that I had indeed 
recorded 32 and 33 Lyncis. 

     I had left home at 7 pm EST, much later than I wanted. If I had 
left before the evening rush hour(s), I would have set up 5 or even 
6 stations.  But with the large south shift, I'm glad I didn't do 
that; what I did was exhausting enough, especially with the 
freeze/thaw cycles I went through (setting up in the cold, then 
warming up during the drive to the next station). At times, I 
thought I should have covered a shorter stretch in the less-covered 
northern zone, and dropped the southern station, but with the large 
south shift, I'm sure glad I kept the southern station.  If we had 
known better the location of the actual path, I would have deployed 
like I did for Brunhild earlier this month, with a fence of stations 
extending south from Cold Harbor. 

        David
___________________

Mark your calendars for the next even brighter asteroidal 
occultation, April 17 at 7:56 pm EDT when 4.2-mag. iota Cancri will 
be occulted by Xanthe (not as large as Palma, but the statistics 
aren't hopeless) in northern Florida. 

Extensive pre-event predictions, maps, finder charts, and observer 
lists are here.  It included full-sky finder charts and descriptions 
to help novice observers locate the target star in a SKY CHARTS 
section.  Standard broadcast radio stations that could be used for 
timing are listed (time-calibrated recordings were made of them).  
Included is an "observing in cold weather" section provided by Guy 
Nason in Ontario.

David Dunham, 2007 Feb. 2, 2h UT
home dunham@starpower.net 301-474-4722 cell 301-526-5590 
office david.dunham@jhuapl.edu 240-228-5609