Occultation light curves

Dave Herald, Murrumbateman, Australia
Derek Breit (USA), David Dunham (USA), Eric Frappa (France), 
Dave Gault (Australia), Tony George (USA), Tsutomu Hayamizu (Japan), 
Brian Loader (New Zealand), Jan Manek (Czech Rep.), Kazuhisa Miyashita (Japan),
Hristo Pavlov (Australia), Steve Preston (USA), Mitsuru Soma (Japan), 
John Talbot (New Zealand), Brad Timerson (USA)

**************************
* Introduction           *
**************************
Lunar occultation light curves have been recorded since the mid-20th century 
using high-speed photomultipliers. Running at high cadence for high angular 
resolution, such recordings were usually made on large telescopes and limited 
to the brighter stars - and were not large in number.

While a small number of video recordings of lunar and asteroidal occultations 
were made from about 1980, they became common from about the year 2000 when 
inexpensive low-light security cameras became available. As of 2016, almost 
all lunar and asteroidal occultation observations are recorded using video, 
with the video recording being measured using software packages such as 
Limovie [http://astro-limovie.info/limovie/limovie_en.html], and Tangra 
[http://www.hristopavlov.net/Tangra3/]. As a result, light curves are now 
routinely generated for almost all lunar and asteroidal occultation 
observations, especially those coordinated through the International 
Occultation Timing Association and related organisations around the world. 
This is resulting in large numbers of occultation light curves being obtained 
each year - albeit with some limitations on time resolution and 
signal-to-noise ratios.

This catalogue is a repository of such light curves. The objective is to make 
these occultation light curves readily available for use when investigating a
star.

******************************
* Nature of the light curves *
******************************
The light curves in this catalogue fall into three distinct categories
* A lunar occultation Disappearance, or Reappearance. The light curve is a 
  single transition between full light and zero light - albeit with the
  potential for one or more intermediate steps caused by companion stars.
  The duration of the light curve is generally only several seconds, but 
  may be longer if a known companion star is present.
* A lunar grazing occultation. These events involve the star clipping along 
  the lunar limb, potentially disappearing and reappearing many times over a 
  period of several minutes.  For these events the motion of the star in a 
  direction normal to the lunar limb is small, whilst the motion parallel to 
  the lunar limb is large. The full duration of the event may be many minutes;
  however the light curve may be split into shorter segments so that the 
  parameters relating to the lunar limb can be better specified.
* An asteroidal occultation. These light curves typically involve a drop in 
  the light curve, followed some time later (typically several seconds, but 
  occasionally over a minute) by a rise in the light curve. Intermediate
  light levels occur when a star is a double star. Binary asteroids can result 
  in more than one drop and rise in the light curve. This can also occur when 
  the observer is close to the edge of the occultation path, and the 
  topography of the asteroid is suitably rough (a 'grazing' occultation).

Apart from intermediate light levels attributable to multiple stars, the light 
curves can exhibit the effects of Fresnel diffraction, and of stellar diameter.

**************************
* Recording equipment    *
**************************
As of 2016, video recordings are mainly made using one or other of the two 
international video standards - NTSC, or PAL. Both NTSC and PAL use an 
interlaced video scan, whereby each frame of the video is comprised of two 
interlaced, time-sequential, fields. The frame rate of an NTSC system is 29.97 
frames/sec (59.94 fields/sec), while that for PAL is 25 frames/sec 
(50 fields/sec). Consistent with broadcast television standards, the majority 
of video cameras used for recording occultations use 8-bit CCD's. However 
some video recordings are made using progressive scan, 12 to 16-bit digital 
video systems.

For fainter objects (in particular, with asteroidal occultations), many 
observers use an 'integrating' video camera. All integrating cameras used 
by observers sum the data from a series of video frames before outputting 
that data - with the output from the video camera being constant for a number 
of frames equal to the integration period. The light curve generated using an 
integrating video camera should normally have only one data point per 
integration interval. While video cameras using a 'running integration' 
scheme are available on the market, their use is discouraged and they are 
rarely used.

Each light curve includes the UTC corresponding to the time at the start of 
the exposure of the first data point in the record. Generally that time has 
been established using the 1PPS output of a GPS receiver - typically linked 
to the video recording via a video inserter. However steps have not been taken 
to ensure absolute accuracy of the start time as specified in this archive. 
Accordingly the start time should be treated as uncertain by an amount 
equivalent to 1.5 data points. 

For some light curves involving asteroidal occultations, the time linking has 
relied on a clock internal to a recording device, with that clock being
calibrated before & after the event. This primarily occurs with observations 
that have been made using unattended equipment, with the calibrations 
occurring up to an hour or more before and after the occultation event. The 
absolute accuracy of the time for the start of such a light curve is likely 
to be as much as a significant fraction of a second; however the relative 
accuracy of the time during the duration of the light curve is essentially 
unaffected. Such situations are identifiable by the presence of several light 
curves of the same star by the same observer from different locations on the 
same date.

It can be assumed that all light curves are 'white light'; observers 
do not routinely insert filters into the optical path. As a result, the optical
pass band for the light curves is that of typical CCD's - good sensitivity 
in the red, and poor sensitivity in the blue. It can generally be assumed 
that the light curves have been generated under circumstances where the camera
response is linear; however the possibility of non-linear response arising 
from saturation cannot be excluded in the case of bright stars.

**************************
* Resolution issues      *
**************************
For lunar occultations, the temporal resolution is governed by a combination 
of the frame (or field) rate of the video recording, and the rate of motion 
of the moon. The typical topocentric motion of the moon is between about 
0.3"/sec and 0.4"/sec. The motion of the lunar limb in a direction normal to 
the star is reduced by the cosine of the difference between the direction of 
motion of the moon and the position angle of the star. As a result, the 
typical rate of motion of the lunar limb normal to the star is in the range 
0.2 to 0.4 "/sec. At video frame rates this provides a spatial resolution of 
about 10 to 20 mas at frame rate, or 5 to 10 mas at field rate. 

In recent years it has been possible to accurately determine the orientation 
of the lunar limb at the point of an occultation, using data from the Japanese 
Kaguya satellite, and more recently the US Lunar Reconnaissance Orbiter - 
Lunar Orbiter Laser Altimeter (LRO-LOLA). The LRO-LOLA data allows the slope 
of the lunar limb to be reliably determined over circumferential distances of 
less than 0.2" in the sky plane.  As a result, all data elements required to 
analyse a lunar occultation light curve are well determined - and are included 
in this archive.

The motion of most asteroids is much less than the moon. As a result, the 
angular resolution attainable at video frame rate is much smaller than for a 
lunar occultation, and is commonly in the range 0.0001" to 0.001". However 
asteroidal occultations frequently involve fainter objects than for lunar 
occultations, and many observers use integrating video cameras to detect 
these fainter occultations; the resolution attainable with an integrating 
camera is reduced in proportion to the number of frames integrated.

Unlike lunar occultations, the orientation of the occulting limb of an 
asteroid relative to the star is generally not well established. 
Furthermore it can generally be assumed that the limb of an asteroid is 
likely to have significant irregularities at scales greater than the 
potential angular resolution attainable, but smaller than the angular 
distance between adjacent observed occultation chords. There is also the 
issue of the rotational orientation of the asteroid differing for observers 
located at different points along the occultation path, placing a limit on 
the accuracy of the limb slope that can be derived from adjacent occultation 
chords. Accordingly, at this time the record does not attempt to specify the 
orientation of the limb of the asteroid at the occultation event.

The light curves in the archive may show the presence of stellar duplicity, 
and occassionally stellar diameter effects and/or Fresnel diffraction. Light 
curves not showing these effects place limits (within the confines of the 
noise in the light curves) on the separation and/or brightness of any 
possible companion to that star on the date of observation, and on the 
diameter of that star.

**************************
* Basis of the catalogue *
**************************
This catalogue contains light curves measured from lunar and asteroidal 
occultations, with the majority coming from lunar occultations. In the main, 
the light curves have been obtained using telescopes having an aperture in 
the range 5 to 40 cm, using 8-bit NTSC or PAL video cameras.

No attempt has been made to limit the light curves to 'the best' of those 
available. All submitted light curves are included, provided they 
(i) show clear evidence of an occultation event, and 
(ii) relevant associated information has been provided. 
As a result, there can be multiple light curves for a single star:
- submitted by the same or different observers on different nights;
- submitted by different observers on the same night, each observing at 
  different sites; or
- submitted by the same observer on the same night, using unattended observing 
  techniques at different sites.
Those light curves can have greatly different signal-to-noise values -
reflecting widely different observational circumstances and equipment used to 
make the recording.

For lunar occultations, the stars fall within a band 13d 20' wide centered on 
the ecliptic. The magnitude limit is highly dependant upon the size of
telescope used, and the phase of the moon at the time of an observation. The 
majority of stars are likely to be brighter than mag 10. Stars fainter than 
mag 12.0 are unlikely to be present.

Asteroidal occultations can occur anywhere in the sky - although the majority 
are within 15 degrees of the ecliptic. While the majority of observed 
occultations involve stars brighter than mag 12, much fainter stars may be 
present.

**************************
* Record content         *
**************************
The information included with each light curve depends on whether it is a 
lunar or asteroidal occultation. The information provided is:

* Date
	Date of the event.
	UTC corresponding to the first point in the light curve (h m s.ss).
	Duration of the light curve (seconds).
	Number of data points in the light curve.
* Star identification, in each of:
		Hipparcos (I/239), Tycho2 (I/259), SAO (I/131A), XZ80Q (I/291), 
		ZC (Robertson's Zodiacal catalogue, US Nautical Almanac, 
		1940USNAO..10..169R),
		UCAC2 (I/289), UCAC4 (I/322A), K2 EPIC ID.
* Observer's longitude (d m s), latitude (d m s), altitude (m), and name.
* Circumstances of the event
- For Lunar occultations:
	Axis angle of the event (the position angle referred to the lunar north 
	  pole).
	Libration in longitude (deg).
	Libration in latitude (deg)..
	Slope of the lunar limb, derived from LRO-LOLA. Positive values correspond 
	to an increase in the PA of the normal compared to the mean limb (deg).
	Normal rate of motion of the star relative to the mean limb ("/sec).
	Contact angle - the angle between the normal to the mean limb, and the 
	  direction of motion of the star (deg). Range is between -90 & +90 for 
	  disappearance events, and between 90 to 180, and -90 to -180. Positive 
	  values apply when the star is northwards of the direction of motion.
	Moon size. The ratio of apparent radius of the moon to its mean radius
	Position angle of the event (deg).
	Cusp angle. The angular distance of the event from the nearer cusp (deg). 
	  The relevant cusp is most commonly indicated as N or S (North or South), 
	  but near full moon the cusp can be E or W (East or West). Positive 
	  values are against the dark limb while negative values are against the 
	  bright limb of the moon.
	Illumination. The percent illumination of the moon.
	Moon altitude. The altitude of the star at the time of the event (deg).
- For asteroidal occultations:
	Asteroid number.
	Asteroid name.
* Data points in the light curve - with the values being normalised to be 
    within the range +/- 9999.

**************************
* References             *
**************************
The lunar occultation observations (but not light curves) are archived in 
Catalogue VI/132B Lunar Occultation Archive. That catalogue provides full 
details of the observation, and is updated on an irregular basis.

Asteroidal occultation observations (but not light curves) are archived in 
NASA Planetary Data System, Small Bodies Node, Asteroid/Dust Archive, as 
Asteroid Occultations. The identifier is EAR-A-3-RDR-OCCULTATIONS-Vxx.x, and 
is located at http://sbn.psi.edu/pds/resource/occ.html
That catalogue provides full details for all asteroidal occultation events, 
and is updated annually.
