Updated: 2003 July 8
     This is now in three parts.  The first goes through the steps 
used to plot the path for a grazing occultation that occurred on
2002 September 29.  The second part is a partially edited and 
illustrated section of the old IOTA occultation manual, including 
the chapters that explain use of the IOTA predictions and ACLPPP 
lunar limb profiles to plot graze paths and set up expeditions.
At the end (third part) you can obtain the prediction files for 
grazes in the northeastern U.S.A. (within 300 miles of Greenbelt, 
Maryland) for 2003.
______________________________________________________________

     Recently, upon request from a local observer, I plotted the 
path for a grazing occultation that occurred in northern Utah on 
September 29th.  The event was clouded out there, but timings of 
that graze, some with video cameras, were obtained of the graze 
farther west in northern California.  The Utah maps are instuctive 
for the overall process of setting up a graze, including applying 
the correction for height above sea level, which is necessary at 
least for heights greater than 200 meters (600 ft.). 

     In order to get the scanned figures to display and print at
the right size, they are all in files created with a relatively 
recent version of MS Word; if you have an older version that can not 
open the files, let me know.

     I plotted the path on the 1:250,000-scale U. S. Geological 
Survey (USGS) map Brigham City that includes the area where the path 
crosses I-15 north of Salt Lake City.  The path crossed Great Salt 
Lake; the next opportunity to reach the path to the west was in 
Nevada several miles west of the Utah line, near Silver Zone Pass, 
where the path crossed I-80.  I don't have the USGS map of that 
area, so I used Delorme's Street Atlas USA version 8 - I highly 
recommend it since it's much more up-to-date than the USGS maps, and 
names all of the streets (that are named).  I plotted the path on a 
1:125,000-scale map generated by Street Atlas USA; that's easier to 
use than with the 1:250,000-scale USGS maps. I also made detailed 
Street Atlas maps of the rural area where the path crossed Riverside 
and Fielding just east of I-15.  Street Atlas doesn't have elevation 
information, so for that, in this case for Nevada where I didn't 
have a USGS map, I obtained a plot of the USGS map of the area from 
www.topozone.com. 

     Also here is a file with the plotting scales that I used, one
for the 1:250,000-scale USGS maps and one for the 1:125,000-scale
Street Atlas maps; I first plotted the sea level limit line on them.
The units of the scale are minutes of arc of latitude, as given in 
the IOTA Grazereg predictions given here for this graze.
The Moon's azimuth is given in the predictions, but you also need 
the azimuth of the limit line, as explained in the 2nd section below 
from the IOTA manual (the terminology of that manual is used here;
you might now read it first before continuing reading this).  You
can obtain the azimuth of the limit just by plotting the limit line 
on a map, and measuring it with a protractor; azimuth is measured 
clockwise starting at 0 at due north, 90 at due east, 180 at due 
south, and 270 at due west.  I used instead a numerical method, 
where Azlim, the azimuth of the limit, is given by

   Azlim = arc tan {[15.0 x cos(phi)]/[phi+7.5 - phi-7.5]}

where phi is the latitude at a specific longitude, phi+7.5 is the
latitude 7.5' east of the specified longitude, and phi-7.5 is the
latitude 7.5' west of the specified longitude.  I used -112 deg. 45'
for the specified longitude; then phi is 41 deg. 35.26' = 41.58767 
deg., cos(phi) = 0.74794 and [15.0 x cos(phi)] = 11.219.  The
difference [phi+7.5 - phi-7.5] needs to be expressed in minutes of 
arc; in this case, it's easiest to use ' (minutes of arc) measured 
from lat. 41 deg. exactly, so then phi+7.5 is 38.17 and phi-7.5 is 
32.33, and their difference is +5.84.  Then, tan {Azlim} is 1.9211
and Azlim = 62.5 deg.  I calcalated D, the angle between the
limit line and the Moon's azimuth, to be 43.0 deg., and then from 
the formula given in the manual below, the elevation shift factor is
TANZ x sine(D) = 0.484.  So if the elevation (h) is about 4400 ft.,
as it is near I-15, the path shifts southeast by 2130 ft. or 0.40 mile.
Near I-80, the elevation is about 5990 ft. and the path shift then
is 0.55 mile.  These "elevation-corrected" limits are plotted on the 
USGS map and 1:125,000-scale Street Atlas map 
parallel with the sea level limit and labelled with the height 
above sea level in ft. 

Then, looking at the predicted profile (ACLPPP),
which I also scanned with my annotations here, connecting the
3's and *'s to form the lunar profile, and drawing horizontal
lines, I determined that the area that should be covered by an
expedition would be from 0.2 mile north of the shifted limit, to
1.2 miles south of it (distances measured perpendicular to the
path).  I just call these extremes "N" and "S", shown on all the
maps, and they can be used to determine the scale of any of
the maps, since the distance between those two lines is
1.4 miles.  I find it convenient to use engineering scales for 
plotting these distances, since at 1:250,000 scale the unit 
division of the 40 scale is one mile, and at 1:125,000 scale the 
unit division of the 20 scale is one mile.  
For one observer, I tried to find the place where the
horizontal line might intersect the actual profile the most times,
and thought that would be at the 0.55 mile south distance,
where, considering the approximate 0.3-mile accuracy of the
Watts datapoints, would last almost a minute before central
graze. It might be even better at the top of the profile, but
you would be risking a miss there; maybe right at the (shifted)
limit line would be good for a 2nd observer.  Any others could
be spaced north and south of the best line by 0.2 to 0.3 miles
or so.

     The basic path was plotted on the maps mentioned above.  I 
then transferred these plots to more detailed plots with street 
names, etc., only transferring the N, S, and 0.55 mile south 
lines, by comparing where they intersected streets and other 
features that are shown on both maps.

     Below is a list of the names of the files (with links) to these 
more detailed maps; I used SA8 for "Steet Atlas USA version 8".  
Maps similar to the SA8 can be obtained on the Web from sites like 
www.mapsonus.com but I have not been able to get lat./long. grids to 
plot on them. 

02929RFM.DOC - Area east of I-15 from SA8, medium view
02929RFD.DOC - Area east of I-15 from SA8, detailed, streets named
02929SZD.DOC - Silver Zone Pass, NV SA8 detailed view

02929GZU.DOC - Topozone map of the Silver Zone Pass graze area
      (no lines are shown on it; it was used only to obtain the
       height above sealevel in the area and to see the topography).

______________________________________________________________

Below are sections on using IOTA's grazing occultation 
predictions extracted from the draft version of the IOTA OBSERVER'S 
MANUAL as of 1994 July 31, by Joan Bixby Dunham, David W. Dunham, 
and Wayne H. Warren Jr., with very limited updates made 2002 Aug. 26
and 2002 Oct. 1.

NOTE:  IN THIS PRELIMINARY VERSION OF THE MANUAL, ALL FIGURES RE-
FERRED TO IN THE TEXT WILL BE FOUND COLLECTED AT THE BACK OF  THE
DOCUMENT.  That's true of the printed version, but not for this
limited on-line version, where links are provided to a few of the
figures only.  THE  FINAL VERSION WILL INCLUDE APPROPRIATE MATERIAL
FROM THE U. S. NAVAL OBSERVATORY'S TOTAL  OCCULTATION  PREDICTION
PAPERS,  SINCE  IOTA  HAS INHERITED THAT SERVICE.   THERE WILL BE
SOME MINOR CHANGES  TO  CHAPTER  6,  BASED  ON  THE  NEW  GRAZING
OCCULTATION PREDICTIONS (PRODUCED BY A NEW COMPUTER PROGRAM) THAT
WAS DISTRIBUTED STARTING IN EARLY 1994.  A FEW SECTIONS WILL
BE EXPANDED AND CHAPTER 13 WILL BE WRITTEN.                      

TABLE OF CONTENTS (Full; only chapters 6 and 7 are included here)
_________________                                                
                                                                 
1.0  INTRODUCTION  . . . . . . . . . . . . . . . . . . . . . .  1
                                                                 
2.0  WHAT IS AN OCCULTATION AND WHY WE OBSERVE THEM  . . . . .  2
2.1  What is an Occultation?   . . . . . . . . . . . . . . . .  2
2.2  What is a Graze?  . . . . . . . . . . . . . . . . . . . .  3
2.3  Why Observe Occultations?   . . . . . . . . . . . . . . .  3
                                                                 
3.0  BEGINNING OBSERVING   . . . . . . . . . . . . . . . . . .  7
3.1  What to Expect  . . . . . . . . . . . . . . . . . . . . .  7
3.2  Event Nomenclature  . . . . . . . . . . . . . . . . . . .  8
  3.2.1  The Lunar Profile   . . . . . . . . . . . . . . . . .  8
3.3  Observing Equipment   . . . . . . . . . . . . . . . . . .  9
  3.3.1  The Telescope   . . . . . . . . . . . . . . . . . . . 10
  3.3.2  The Radio   . . . . . . . . . . . . . . . . . . . . . 12
  3.3.3  The Tape Recorder   . . . . . . . . . . . . . . . . . 13
  3.3.4  The Stopwatch   . . . . . . . . . . . . . . . . . . . 13
  3.3.5  Other Equipment   . . . . . . . . . . . . . . . . . . 14
  3.3.6  Equipment List  . . . . . . . . . . . . . . . . . . . 14
3.4  Making Scientifically Useful Occultation Timings  . . . . 16
3.5  Reporting Observations  . . . . . . . . . . . . . . . . . 17
                                                                 
4.0  ACCURATE TOPOCENTRIC POSITION DETERMINATION   . . . . . . 18
4.1  Definition  . . . . . . . . . . . . . . . . . . . . . . . 18
4.2  Accuracy Requirements   . . . . . . . . . . . . . . . . . 19
4.3  Measuring Coordinates   . . . . . . . . . . . . . . . . . 19
  4.3.1  Coordinates from GPS Measurements   . . . . . . . . . 21
4.4  Example of Coordinate Determination   . . . . . . . . . . 22
  4.4.1  Position Determination  . . . . . . . . . . . . . . . 22
  4.4.2  Errors  . . . . . . . . . . . . . . . . . . . . . . . 23
  4.4.3  Numerical Example   . . . . . . . . . . . . . . . . . 24
                                                                 
5.0  TOTAL OCCULTATIONS  . . . . . . . . . . . . . . . . . . . 25
5.1  Observing Techniques  . . . . . . . . . . . . . . . . . . 25
5.2  Techniques for Observing Reappearances  . . . . . . . . . 26
5.3  Predictions   . . . . . . . . . . . . . . . . . . . . . . 28
                                                                 
6.0  GRAZING OCCULTATION PREDICTIONS   . . . . . . . . . . . . 29
6.1  Grazing Occultation Limit Predictions   . . . . . . . . . 29
  6.1.1  Heading Data  . . . . . . . . . . . . . . . . . . . . 29
  6.1.2  Column Data   . . . . . . . . . . . . . . . . . . . . 32
  6.1.3  Ending Data   . . . . . . . . . . . . . . . . . . . . 34
  6.1.4  Elevation Correction  . . . . . . . . . . . . . . . . 35
6.2  Profile Predictions   . . . . . . . . . . . . . . . . . . 35
  6.2.1  Profile Explanation   . . . . . . . . . . . . . . . . 36
  6.2.2  Limit Prediction and Profile Use  . . . . . . . . . . 40
    6.2.2.1  Topographic Maps  . . . . . . . . . . . . . . . . 40
    6.2.2.2  Graze Path Plotting   . . . . . . . . . . . . . . 40
    6.2.2.3  Profile Use   . . . . . . . . . . . . . . . . . . 42
                                                                 
7.0  ORGANIZING GRAZING OCCULTATION EXPEDITIONS  . . . . . . . 47
7.1  Preparation   . . . . . . . . . . . . . . . . . . . . . . 47
7.2  Site Selection  . . . . . . . . . . . . . . . . . . . . . 47
  7.2.1  Observer Notification and Preparation   . . . . . . . 49
7.3  Expedition Reporting  . . . . . . . . . . . . . . . . . . 49
7.4  Approximate Reduction and Shift Determination   . . . . . 50
  7.4.1  Correction of Profile to Actual Observing Location    50
  7.4.2  Correcting the Sea Level Limit Prediction for Ele-      
   vation  . . . . . . . . . . . . . . . . . . . . . . . . . . 53
  7.4.3  Calculating the Time of Central Graze   . . . . . . . 54
  7.4.4  Plotting the Observations on the Predicted Profile    55
  7.4.5  Calculating the Shift Value   . . . . . . . . . . . . 55
                                                                 
8.0  ACCURATE TIMING   . . . . . . . . . . . . . . . . . . . . 57
8.1  Methods of Timing   . . . . . . . . . . . . . . . . . . . 57
  8.1.1  Stopwatch method  . . . . . . . . . . . . . . . . . . 57
  8.1.2  Tape Recorder Method  . . . . . . . . . . . . . . . . 60
  8.1.3  Eye and Ear Method  . . . . . . . . . . . . . . . . . 61
  8.1.4  Alternate Methods for Limited Equipment   . . . . . . 62
    8.1.4.1  AM Radio as a Time Standard   . . . . . . . . . . 62
    8.1.4.2  Recording Assistant   . . . . . . . . . . . . . . 62
    8.1.4.3  CB Radio in Panic Mode  . . . . . . . . . . . . . 63
  8.1.5  Photoelectric Method  . . . . . . . . . . . . . . . . 63
  8.1.6  Video   . . . . . . . . . . . . . . . . . . . . . . . 64
8.2  Determining Personal Equation and Timing Accuracy   . . . 65
8.3  International Time Standards and Time Signals   . . . . . 67
  8.3.1  Time Scale Definitions  . . . . . . . . . . . . . . . 68
  8.3.2  Coordinated Universal Time  . . . . . . . . . . . . . 70
  8.3.3  Time Signals  . . . . . . . . . . . . . . . . . . . . 71
  8.3.4  Radio Propagation Delay   . . . . . . . . . . . . . . 72
  8.3.5  Time Delay of Sound   . . . . . . . . . . . . . . . . 73
  8.3.6  Signal Reception Quality  . . . . . . . . . . . . . . 73
  8.3.7  Timings Without Shortwave Time Signals  . . . . . . . 74
                                                                 
9.0  ASTEROIDAL OCCULTATIONS   . . . . . . . . . . . . . . . . 75
9.1  Predicting Asteroidal Occultations  . . . . . . . . . . . 75
9.2  Asteroid Occultation Prediction Updates   . . . . . . . . 76
9.3  Observing Occultations by Asteroids   . . . . . . . . . . 76
                                                                 
10.0  OTHER OCCULTATION EVENTS   . . . . . . . . . . . . . . . 78
10.1  Predictions of Special Events  . . . . . . . . . . . . . 78
10.2  Observing Special Events   . . . . . . . . . . . . . . . 79
10.3  Occultations by Comets   . . . . . . . . . . . . . . . . 79
                                                                 
11.0  SOLAR ECLIPSE OBSERVATIONS   . . . . . . . . . . . . . . 80
11.1  Solar Eclipse Phenomena near the Edge of the Path of To-   
 tality  . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
11.2  Eye and Equipment Safety   . . . . . . . . . . . . . . . 81
11.3  Observing Techniques   . . . . . . . . . . . . . . . . . 82
  11.3.1  Projection   . . . . . . . . . . . . . . . . . . . . 82
    11.3.1.1  Visual Observation of the Projected Image  . . . 83
    11.3.1.2  Photography of the Projected Image   . . . . . . 83
    11.3.1.3  Video Recording of the Projected Image   . . . . 84
11.4  Direct Photography   . . . . . . . . . . . . . . . . . . 84
  11.4.1  Photographing with a Movie Camera  . . . . . . . . . 86
  11.4.2  Photographing with a Video Camera  . . . . . . . . . 87
11.5  Timing   . . . . . . . . . . . . . . . . . . . . . . . . 87
11.6  Site Selection   . . . . . . . . . . . . . . . . . . . . 87
11.7  Reporting Observations   . . . . . . . . . . . . . . . . 88
11.8  Observing Hints and Suggestions  . . . . . . . . . . . . 88
                                                                 
12.0  PREDICTING OCCULTATIONS  . . . . . . . . . . . . . . . . 90
12.1  Lunar Total Occultations   . . . . . . . . . . . . . . . 90
12.2  Lunar Grazing Occultations   . . . . . . . . . . . . . . 90
12.3  Lunar Librations   . . . . . . . . . . . . . . . . . . . 91
12.4  Eclipses   . . . . . . . . . . . . . . . . . . . . . . . 91
12.5  Asteroidal Occultations  . . . . . . . . . . . . . . . . 91
12.6  Planetary Occultations   . . . . . . . . . . . . . . . . 92
                                                                 
13.0  ADVANCED OBSERVING TECHNIQUES  . . . . . . . . . . . . . 93
                                                                 
APPENDIX A. OCCULTATION SERVICES AND SOURCES OF FURTHER INFOR-   
 MATION  . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
A.1  IOTA  . . . . . . . . . . . . . . . . . . . . . . . . . . 94
A.2  International Lunar Occultation Centre  . . . . . . . . . 94
A.3  Observer's Handbook   . . . . . . . . . . . . . . . . . . 95
A.4  Sky and Telescope, Astronomy Magazines  . . . . . . . . . 95
                                                                 
APPENDIX B. REFERENCES   . . . . . . . . . . . . . . . . . . . 96
                                                                 
APPENDIX C. PLOTTING SCALES  . . . . . . . . . . . . . . . . . 98
                                                                 
APPENDIX D. EQUIPMENT CHECK LIST   . . . . . . . . . . . . . . 99
                                                                 
APPENDIX E. DARK LIMB AND TERMINATOR DATA  . . . . . . . . .  101
E.1  Terminator Dark Limb Separations  . . . . . . . . . . .  101
E.2  Elongation Data   . . . . . . . . . . . . . . . . . . .  104
                                                                 
APPENDIX F. REPORT FORMS   . . . . . . . . . . . . . . . . .  106
F.1  ILOC material, where to get forms, blank forms  . . . .  106
F.2  Asteroid Events   . . . . . . . . . . . . . . . . . . .  106
F.3  Need to add explanation of IOTA terms at the bottom   .  106
                                                                 
APPENDIX G. SAMPLE GRAZE PROFILE   . . . . . . . . . . . . .  107
                                                                 
APPENDIX H. FIGURES  . . . . . . . . . . . . . . . . . . . .  108
H.1  Figure Captions   . . . . . . . . . . . . . . . . . . .  108
                                                                 
6.0  GRAZING OCCULTATION PREDICTIONS                             
____________________________________                             
                                                                 
Grazing  occultation predictions and predicted lunar profiles are
distributed to IOTA members by volunteers,  known  as  computors,
who  prepare  the  predictions from data that, through 1993, were
provided by the United States Naval Observatory.   From 1994  on,
IOTA  plans  to  generate  different  basic  graze  data  for the
computors, who will still calculate and distribute detailed  pre-
dictions  to  IOTA members.   These predictions give the expected
paths for northern or southern limit grazes within a travel range
the observer has selected.  The grazes are rated based on the ex-
pected ease or difficulty of observation, and information is pro-
vided about the star.  Each observer is provided with  a  summary
page  listing  all  predicted events in the time period for which
individual limit predictions are included.  A summary page is il-
lustrated in Figure 6-1 and a recent example is at the top of the
prediction file given here.
The  profile  predictions,  prepared  from  Watts  charts of  the 
marginal zone of the Moon, give the observer an indication of the 
distance  to each  side of  the limit  line where multiple events 
will be seen.                              
                                                                 
6.1  GRAZING OCCULTATION LIMIT PREDICTIONS                       
__________________________________________                       
                                                                 
The grazing occultation limit prediction data are provided as ta-
bles  of  longitude,  latitude,  and time for the predicted path,
with the Moon's altitude and azimuth and the Sun's altitude  also
indicated.    Each  graze prediction has a heading section, which
gives information on the star and the  graze  rating.    The  lo-
cations of other observers selected to receive a particular graze
prediction  are  given at the end of the tabular data. An example
of a  graze limit prediction is given  in Figure 6-2 and a recent 
example is here.  Starting  in  1994,
IOTA  started  using  a  different  computer  program  for  graze
predictions. The format, illustrated with the example at the link
above (The new program is called Grazereg,  by Eberhard Riedel in
Munich, Germany),  is a  little different  from  the  description
below of the  old (pre-1994)  predictions,  but most of the  same 
information is given.  
                                                                 
6.1.1  HEADING DATA                                              
___________________                                              
                                                                 
LINE 1  These include the name; super standard station letter(s);
        city; country, state, or  province;  and  maximum  travel
        radii  of the observer.  For the purpose of graze predic-
        tions, much of the world has  been  divided  into  super-
        standard  station  regions,  each 1000 miles (1609 km) in
        diameter. Different computors  (volunteers  who  run  the
        computer programs to do the actual predictions) calculate
        and  distribute  predictions  for  the different regions.
        Observers whose maximum travel radii cover more than  one
        super-standard  region  receive more than one set of pre-
        dictions, one from each region their maximum  travel  ra-
        dius covers.                                             
                                                                 
LINE 2  The  distance  in  miles of the closest point in the pre-
        dicted limit to the posiiton specified by the observer as
        his location, the Universal Time (UT) of central graze at
        that point, and the graze rating are given  here.    When
        the  program computes these quantities, it terminates the
        path at low Moon altitude or when the Sun  is  above  the
        horizon.    It may be that the actual path will be closer
        to the observer than the number specified as the  closest
        point,  but low altitude or daylight may make observation
        impossible at the true closest point.   Twilight and  low
        Moon  altitude are not considered in computing the rating
        (except for  spectacular  events),  but  interference  by
        sunlit lunar features and daylight are considered.       
                                                                 
LINE 3  The third line is normally blank, but sometimes a special
        message  appears  here,  giving  information  such as the
        computor's address, spectroscopic binary data, or a lunar
        eclipse message.                                         
                                                                 
LINE 4  This line gives the star's name (if any),  Durchmusterung
        (BD  or  CD)  number, USNO reference number (X, ZC, K, C,
        etc.),  Smithsonian  Astrophysical  Observatory   Catalog
        (SAO)  number,  and  visual magnitude, followed by the UT
        date.  The USNO reference number and the SAO  number  are
        the  numbers  that  should  be entered on observation re-
        ports.  SAO numbers less than 17 are not  true  SAO  num-
        bers,  but  are  error  codes  for  stars  in  the  Third
        Astronomische Gesellschaft Katalog (AGK3) that are not in
        the SAO catalog.  The name includes a proper name,  Greek
        letter,  Flamsteed number, variable-star designation, and
        other catalog numbers, in that order of  preference,  and
        depending  on  which,  if any, are available.   The other
        catalog numbers, indicated by the letters B., H1., G. and
        H.  (for Bode, Heis, Gould, and  Hevelius,  respectively)
        should  not  be  confused  with  the  more  commonly used
        Flamsteed numbers.  The number immediately following "BD"
        or "CD" is the "zone" number, and is the star's  approxi-
        mate declination in degrees.                             
                                                                 
LINE 5  The line starts with the percent of the Moon's disk illu-
        minated  by  the  Sun,  where  0 is new moon, 100 is full
        moon, 50 is first or last quarter, 1 to  49  is  crescent
        moon, and 51 to 99 is gibbous moon.  If the Moon's appar-
        ent  diameter is considered to be 100 units, the termina-
        tor crosses  the  Moon's  equator  at  a  point  "percent
        sunlit"  units  measured  along  the  terminator from the
        bright limb.  The equator here is not the  Moon's  actual
        equator,  but  is  usually very close to it; it is really
        the diameter line through the Moon's center  pointing  in
        the  direction of the Sun.  The term WAXING following the
        percent sunlit shows that the percent sunlit is  increas-
        ing  (between new moon and full moon), while WANING indi-
        cates decreasing percent sunlit (full moon to new  moon).
        The  term  ECLPNG  indicates  that  a lunar eclipse is in
        progress.  During a lunar eclipse, the percent sunlit  is
        the  percent  of  the Moon's diameter not in the umbra at
        the central graze time given in the second line, so  that
        0  would  imply  totality.   Note that the percent sunlit
        changes rapidly during the partial phases of an  eclipse.
        The position angle of cusp is geocentric and only approx-
        imate;  it  is  meaningless during an eclipse.  DELTAT is
        the difference, Ephemeris Time minus  Universal  Time,  a
        quantity with which observers are likely never to be con-
        cerned.    Finally,  it  is  noted  whether the path is a
        northern or southern occultation limit.                  
                                                                 
LINE 6  This line gives the error of the star's declination,  and
        the  phase  of the Moon.  The PROBABLE ERROR is a measure
        of the uncertainty in the star's position.   There  is  a
        50%  chance  that  the  star will actually be between the
        star's catalog declination + the error  and  the  catalog
        declination ¯ the error.  It should be used with the pre-
        dicted  profile  for  determining  the  distance from the
        limit line for positioning observers.                    
                                                                 
LINE 7  The SPECTRAL CLASS indicates the star's color.  Stars  of
        class  O  and  B  are blue; A, bluish-white; F, white; G,
        yellowish-white; K, orange; M, N, S, and C, red.  The Sun
        (and therefore the Moon)  is  G2,  yellow.    Bright-limb
        grazes are easier to see if there is good color contrast,
        as for B and M stars.  B, A, and F stars have the highest
        probability  of  being double.  The POSITION SOURCE gives
        the catalog from which the star's position and other  in-
        formation were taken.  The most reliable catalogs are the
        Fourth  Fundamental Catalog (FK4), its supplement (FK4S),
        and Washington N30 Catalog  (N30).    Less  reliable  are
        Robertson's  Zodiacal  Catalog  (ZC), the Yale Zone Cata-
        logue (Yale University), and the  AGK3,  which  sometimes
        have  accumulated  proper-motion errors amounting to well
        over 1",  in  spite  of  the  stated  probable  error  of
        declination.    The Albany General Catalog (GC) positions
        are often poor, due to the early epoch  of  the  catalog,
        and  errors  in  excess of 2" occur.  With some work, im-
        proved positions for GC and ZC stars fainter than  magni-
        tude  5.5  can  be obtained by consulting other catalogs,
        usually Yale.  For stars north of declination ¯4 degrees,
        all available accurate astrometric catalogs were  used  ,
        when  the XZ catalog was created, to obtain the best pos-
        sible positions and proper motions  on  the  FK4  system.
        The source for these stars is identified as "XZ".  A pre-
        liminary  version  of  the  Zodiacal Zone Catalog (ZZ87),
        based on observations made around 1980 combined with  re-
        measured  and reduced 1930's data for proper motions, has
        been used for most SAO stars in the XZ.  These  FK4-based
        (B1950.0)  catalogs  will  soon  be replaced by FK5-based
        (J2000.0) catalogs, such as the FK5 itself and its exten-
        sions, the Positions and Proper Motions (PPM) catalog  of
        the Astronomisches Rechen-Institut in Heidelberg, and the
        Astrographic  Catalog Reference Stars (ACRS) catalog pre-
        pared at the U. S. Naval Observatory.   Eventually,  more
        accurate  data  will  become  available from the European
        Space Agnecy's Hipparcos satellite.  IOTA has a  list  of
        program  stars, including the brighter XZ objects, in the
        Hipparcos Input Catalogue (HIP), from which the observing
        program is taken.  The PREDICTION BASIS is the USNO lunar
        ephemeris used in determining the prediction.    The  ob-
        server will rarely need to be concerned with it.         
                                                                 
LINE 8    Extra lines in the heading usually are for double-star 
        data,  including  the separation(s) and position angle(s)
        for secondary and possible  tertiary  (third)  components
        from  the  primary.    If  the star is listed in Aitken's
        double-star catalog, this is indicated, since his  desig-
        nations  are  often  used  in  double-star work.   If the
        statement "THE POSITION SOURCE MUST BE CONSULTED FOR  PO-
        SITION  USED"  appears,  the  position of the primary has
        usually been used for the prediction,  unless  the  sepa-
        ration is less than about 3", and the secondary is nearly
        as  bright as the primary.  In this case, a mean position
        ( probably the center of light of the  system)  has  been
        used.  Offsets for mean position are computed by the pro-
        gram  that produces the profiles.  During lunar eclipses,
        a message describing the umbral distance is printed.     
                                                                 
6.1.2  COLUMN DATA                                               
__________________                                               
                                                                 
Twelve columns of data are given at regualar intervals of  longi-
tude  along  the graze path.   The longitude is measured westward
(negative numbers indicate east longitude) from Greenwich.       
                                                                 
WEST LONGITUDE  The longitude in degrees and decimals of degrees.
                The intervals are usually 0.125 deg.in longitude,
                or 7.5', so they are at the margins of  7.5' USGS
                topographic maps.  Data at 2.5', 5.0', and 10.0' 
                intervals  are  also  available.    Below are    
                listed the fractions  of  degrees  at  0.125  and
                their equivalents in minutes and seconds for ref-
                erence with the topographic maps.                
                                                                 
                +--------------+-------------+                   
                |    0.875     |   52' 30"   |                   
                |    0.750     |   45  00    |                   
                |    0.625     |   37  30    |                   
                |    0.500     |   30  00    |                   
                |    0.375     |   22  30    |                   
                |    0.250     |   15  00    |                   
                |    0.125     |   07  30    |                   
                |    0.000     |   00  00    |                   
                +--------------+-------------+                   
                                                                 
                 Table 1.  Fractional  Degree                    
                           Equivalents                           
                                                                 
NORTH LATITUDE  The degrees  and  minutes  of  geodetic  latitude
                where  the  predicted sea level limit crosses the
                longitude meridian of WEST LONGITUDE.  Minutes of
                arc are probably easiest to use with  topographic
                maps;  if  desired, the seconds of arc can be de-
                termined by multiplying the decimal part  of  the
                minutes by 60.                                   
                                                                 
UNIVERSAL TIME  The  coordinated  universal time (UTC) of central
                graze as seen from the  longitude  and  latitude.
                This  is the time when the star is closest to the
                center of the Moon as seen from that location.   
                                                                 
MOON ALTITUDE   Altitude of the star being occulted.   The  alti-
                tude is zero at the horizon and 90d at the zenith
                and  is equal to 90d minus the zenith angle.  At-
                mospheric refraction is  not  considered  in  the
                calculation of the altitude.                     
                                                                 
MOON AZIMUTH    Azimuth  of the star being occulted.  The azimuth
                is measured eastward from due north, so that  90d
                is  due  east, 180d is due south, and 270d is due
                west.                                            
                                                                 
TANZ            The tangent of the zenith angle of the star.   It
                is  also  the  cotangent  of MOON ALTITUDE and is
                used with MOON AZIMUTH to compute the amount  the
                sea level limit must be moved for heights signif-
                icantly  above  mean sea level.   This correction
                should be performed when the observer's height is
                more than  200  meters  (aproximately  600  feet)
                above sea level.  This is explained in "Elevation
                Correction" on page 35.                          
                                                                 
SUN ALTITUDE    The Sun's altitude in degrees and tenths of a de-
                gree.    It is negative when the Sun is below the
                horizon and positive when above. Atmospheric  re-
                fraction is not taken into account, so 0.6 deg.  
                should  be added when the altitude is within a   
                degree of the horizon.  Astronomical twilight be-
                gins when the Sun reaches -18d, nautical twilight
                begins when the Sun's altitude is -12d, and civil
                twilight when the Sun reaches -6d.               
                                                                 
PA OF GRAZE     The  position  angle  of central graze in degrees
                and tenths, measured eastward  along  the  Moon's
                limb from north.  North on the Moon's disk is de-
                fined  to  be  from  the right ascension meridian
                passing from the center of the Moon to  the  cur-
                rent (apparent) North Celestial Pole (NCP).      
                                                                 
CUSP ANGLE      The  angle  measured  in  degrees around the limb
                from the cusp to  the  point  of  central  graze,
                where  N  or  S  indicate  whether it is from the
                north or from the south cusp,  respectively.    A
                negative  number indicates that the point of cen-
                tral graze is on the sunlit  limb  of  the  Moon,
                while  a positive number indicates the dark limb.
                The cusp is the intersection  of  the  terminator
                with  the  Moon's  limb, 90þ around the limb from
                the direction to the Sun from the center  of  the
                Moon.    High  mountains  beyond this theoretical
                cusp  often  catch  sunlight.      During   lunar
                eclipses,  the  cusp  angle is meaningless and is
                replaced by the UMBRAL DISTANCE, the distance  of
                the  star from the center of the umbra, expressed
                as a percent of the radius of the umbra.  U  dis-
                tinguishes  the  umbral  distance from the N or S
                given with cusp angles.                          
                                                                 
6.1.3  ENDING DATA                                               
__________________                                               
                                                                 
Following the columns of data are statements indicating the  ver-
sion  of the prediction program used, the prediction data source,
and the name of the computor who ran the program.   The last  in-
formation  given  is  a  result  of the observer scan, indicating
which other observers have been selected to receive these predic-
tions, if any.   In the scan,  the  observer's  specified  travel
radii are given in miles after the latitude.  An asterisk follow-
ing the spectacular radius signifies those who expect to organize
expeditions  more  often  than join expeditions set up by others.
The super standard station, in which the observer's  position  is
located,  is  given after his name.  The time of closest approach
is in hours and decimals of an hour.                             
                                                                 
Some, but not all, versions of the graze prediction programs gen-
erate a one-page summary of all grazes.   The observer's  station
coordinates  are given in the heading.  The summary includes cir-
cumstances at the point of closest approach, including the longi-
tude and latitude of the closest point given in  the  main  list.
Some  summaries give the bearing of the limit (the azimuth of the
motion of the shadow), the double-star code  (under  "DBL"),  and
the faintest magnitude for variable stars.                       
                                                                 
6.1.4  ELEVATION CORRECTION                                      
___________________________                                      
                                                                 
MOON  AZIMUTH  and  TANZ can be used to make corrections for ele-
vations above sea level.  If the elevation above sea level  is  h
(in  feet  or  meters), the magnitude of the correction to be ap-
plied, d, is given by the formula                                
                                                                 
                            d = (TANZ) h                     
                                                                 
where d is  applied  in the direction of the  Moon's azimuth; see 
the small diagram  (Fig. 6-3) in the  upper right  corner of  the 
graphic shown here.  If the limit is  
plotted and  its direction or azimuth  is measured, the distance, 
x, that the limit should  be  shifted,  x  measured perpendicular 
to the limit, is given by the formula:             
                                                                 
                     x = sin (D) d = sin (D) (TANZ) h            
                                                                 
where  D is the difference of the Moon's and the limit's azimuth.
This is illustrated in the map  diagram in the upper left part of
the graphic shown here (x is the distance
from I to H shown on the diagram).  
When the altitude of the star is small, TANZ is large, and  D  is
small (the azimuth of the limit is nearly equal to MOON AZIMUTH).
For  grazes  north  of latitude +31d, note that shifts are always
southward (unless the elevation is below sea level).             
                                                                 
6.2  PROFILE PREDICTIONS                                         
________________________                                         
                                                                 
Graze observers are provided with computer-generated profiles  of
the  limb  of  the  Moon as it is predicted to be observed at the
time of the grazing occultation.  The data used  to  compute  the
profiles  are  the  Watts'  Marginal Zone of the Moon in computer
form, with additional corrections determined from occultation ob-
servations.  These profiles are not exact; they are only  a  pre-
diction  of  what  the  observer  might  see.    They are used by
observers  to  determine  where  the  best  region  is   in   the
occultation  path  for observing.  The profiles are for the point
on the limit line closest to the observer.   The Watts  angle  of
central  graze,  position  angle of the graze, and cusp angle are
all shifted from the longitude and latitude printed on  the  pro-
file  to  the  longitude and latitude of the closest point on the
limit line.  The time of central graze printed under the  profile
is  not  shifted, and should not be used.  The central graze time
for the intended observation area should  be  obtained  from  the
limit predictions.  A recent example of IOTA's ACLPPP profiles is
here.  The format is almost identical to that described
below.   These profiles are now mostly quite accurate, especially
if they are based on observed data  (profile points coded as 3 or
4).  The  profiles that accompany the  Grazereg limit predictions
are  almost  as accurate  but should  be used  only if  an ACLPPP 
profile is not available;  ACLPPP provides  information about the
"worst" terminator (see below), and double and triple stars, that
is not in the Grazereg profiles.
                                                                 
6.2.1  PROFILE EXPLANATION (for ACLPPP profiles)                 
__________________________                                       
                                                                 
PROFILE HEADING DATA  The  following  information is found in the
                      heading:                                   
                                                                 
                      LINES 1-2  These are the scale in Watts an-
                                 gles.    The values on the scale
                                 are printed in one-degree inter-
                                 vals from  the  Watts  angle  of
                                 central graze.                  
                                                                 
                      LINE 3     This  is  the  time from central
                                 graze in  one-minute  intervals.
                                 Vertical  bars are generated for
                                 each minute  through  the  plot,
                                 with  the  one for central graze
                                 so labeled.    Negative  numbers
                                 indicate  minutes before central
                                 graze, and positive numbers min-
                                 utes after central graze.       
                                                                 
PROFILE PLOT          Horizontal bars are drawn across  the  plot
                      at regular intervals to help in scaling the
                      distance  from the predicted limit.  One of
                      the lines is the predicted  limit,  and  is
                      labeled  as  "PREDICTED LIMIT" on the right
                      side, and "0" on the left side.  The verti-
                      cal scale on the right side is  the  number
                      of  miles  or  kilometers  from  the limit,
                      while the scale on the left side is seconds
                      of arc from the limit.  A negative value is
                      south of the limit and a positive value  is
                      north of the limit.                        
                                                                 
                      The  actual  profile  data  are a series of
                      letters, numbers, and  asterisks  that  can
                      look  busy and bewildering.  Some observers
                      have  found  that  drawing  smooth   curves
                      through the points for the limb and for the
                      predicted profile help in understanding the
                      plot.    There  are at least two curves re-
                      presented on each plot, and sometimes more.
                      Each plot has a curve for the  smooth  mean
                      limb  of  the  Moon, and a more jagged plot
                      for the predicted profile.    In  addition,
                      the  terminator  may appear on the plot, if
                      it is near the central graze.  If the  star
                      is  double, and both components will graze,
                      the profile for the secondary  (and  terti-
                      ary,  if  there is one) component will also
                      be provided.                               
                                                                 
                      The codes for  the  limbs  and  terminators
                      are:                                       
                                                                 
                      D  dark limb of the Moon                   
                                                                 
                      B  bright limb of the Moon                 
                                                                 
                      T  terminator                              
                                                                 
                      W  "worst"   terminator,   where   two-mile
                         (3-km) high lunar mountain peaks can  be
                         sunlit.  Areas enclosed by W's will usu-
                         ally  be sunlit at the south limb, where
                         high mountains are common, and will usu-
                         ally be dark at  the  relatively  smooth
                         north limb                              
                                                                 
                      The codes for the profile points are:      
                                                                 
                      *  good limb correction, typically accurate
                         to #0                                   
                          . overlay '"'                          
                         15                                      
                                                                 
                      1  fair limb correction, accurate to #0    
                          '.' overlay '"'                        
                         3                                       
                                                                 
                      2  meaningless  limb correction, either ex-
                         treme librations or in the  Cassini  re-
                         gion  (see  "The  Lunar Profile" on page
                         8).                                     
                                                                 
                      3  good limb correction from previously ob-
                         served graze data, accurate to #0       
                          '.' overlay '"'                        
                         4                                       
                                                                 
                      4  poor limb correction from previously ob-
                         served graze data, accurate to #1"; most
                         of the Cassini regions have been crudely
                         "mapped"   with   previously    observed
                         grazes,  so 3's and 4's usually dominate
                         the profile when a graze occurs in these
                         regions                                 
                                                                 
                      5  good limb correction with  an  empirical
                         correction applied (*[or 0] + 5)        
                                                                 
                      6  fair  limb  correction with an empirical
                         correction applied (1 + 5)              
                                                                 
                      7  meaningless limb correction with an  em-
                         pirical correction applied (2 + 5)      
                                                                 
                      P  shifted limb of the primary component of
                         a multiple star (when the star is not at
                         the  position  used for the limb predic-
                         tions, which is often the  case  when  a
                         center-of-light,  or  mean  position, is
                         used)                                   
                                                                 
                      S  shifted limb of the secondary  component
                         of a multiple star                      
                                                                 
                      R  shifted  limb  of the tertiary component
                         of a multiple star                      
                                                                 
                      When drawing  curves  through  the  plotted
                      points, the following groups should be con-
                      nected together.  A different color pen for
                      each group makes the profile more readable.
                                                                 
                      B AND T  enclose bright area of the Moon   
                                                                 
                      D        encloses dark mean limb           
                                                                 
                      W        encloses  area  where sunlit peaks
                               may exist and cause observing dif-
                               ficulties ("worst" terminator)    
                                                                 
                      *,1-7    the predicted limb for  mean  star
                               position                          
                                                                 
                      P        the  predicted  limb for a primary
                               star not at the mean star position
                                                                 
                      S        the predicted limb for a secondary
                               component                         
                                                                 
                      R        the predicted limb for a  tertiary
                               component                         
                                                                 
                      At  the  bottom  of  the  profile are seven
                      lines of additional information  about  the
                      profile and the star.                      
                                                                 
                      LINE 1     The   date,  time  and  latitude
                                 libration of the graze is given.
                                                                 
                      LINE 2     The star number in the ZC  or  X
                                 catalog, the version of the USNO
                                 profile  prediction program that
                                 generated the data for the plot,
                                 and the longitude  libration  of
                                 the  graze  are given.  The time
                                 is for the "standard"  longitude
                                 given in line 5.                
                                                                 
                      LINE 3     This   line   gives   the  limit
                                 (northern,  southern)   of   the
                                 graze, the vertical profile cor-
                                 rection  (VPC) in seconds of arc
                                 (north is positive), and the po-
                                 sition angle of the graze at the
                                 "standard"  longitude  given  on
                                 the 5th line.                   
                                                                 
                      LINE 4     The Watts angle of central graze
                                 (for  the  point  in  the  limit
                                 closest  to the observer), graze
                                 height (of the  predicted  limit
                                 from the mean limb in seconds of
                                 arc,  with positive values indi-
                                 cating a  shift  away  from  the
                                 center  of  the  Moon), the cusp
                                 angle, and the name of the  per-
                                 son  running the profile program
                                 are given.                      
                                                                 
                      LINE 5     The longitude for the basic pre-
                                 diction data (this either is the
                                 closest point in  the  predicted
                                 limit,  or  west  of the closest
                                 point), the  horizontal  profile
                                 scale  (HPS)  in minutes of time
                                 per degree of Watts  angle,  the
                                 position  angle for the point in
                                 the limit  closest  to  the  ob-
                                 server,  and the person for whom
                                 the  profile  was  produced  are
                                 given.                          
                                                                 
                      LINE 6     The  latitude for the basic pre-
                                 diction,  the  vertical  profile
                                 scale  (VPS)  in  seconds of arc
                                 per mile or per  kilometer,  the
                                 distance  of the observer to the
                                 limit line, and  the  observer's
                                 home location are given.        
                                                                 
                      LINE 7     The  empirical  corrections  ap-
                                 plied are identified.           
                                                                 
                      LINES 8-9  Additional information is  given
                                 if  the  star is multiple.   The
                                 type (double, triple), the  dis-
                                 tance  of  the  primary from the
                                 mean  position,  the  magnitude,
                                 separation,  position angle, and
                                 the  vertical   and   horizontal
                                 shift  in  the  profile for each
                                 component are listed.   Vertical
                                 shifts  are  given  in miles (or
                                 kilometers) and seconds of  arc,
                                 horizontal  shifts in time.  In-
                                 formation  is  given  for compo-
                                 nents that do not  show  on  the
                                 plot.                           
                                                                 
6.2.2  LIMIT PREDICTION AND PROFILE USE                          
_______________________________________                          
                                                                 
Observers  can use the grazing occultation predictions, described
later in this section, and the predicted  profiles  to  calculate
the  approximate sequence of events that will be visible from any
point in the vicinity of the northern or  southern  limit  of  an
occultation.  Therefore, observers can position themselves in the
best  locations  near  the  limit to see the most spectacular se-
quence of events to obtain the most useful data.                 
                                                                 
6.2.2.1  TOPOGRAPHIC MAPS                                        
                                                                 
First, the observer should examine a map of his/her state or  re-
gion  using  the computer predictions to determine an approximate
location where the graze might be observed.  One or two  alterna-
tive  locations  might  be  selected in case bad weather or other
circumstances prevent observation at the first  location.    Aer-
onautical  charts  obtainable  at  any airport, topographic index
maps, atlas maps, and even the rare road maps with  latitude  and
longitude  lines can be used for the selection of approximate lo-
cations. The 1:250,000-scale topographic maps available for  most
areas  show  virtually  all roads, cover a large area, and can be
used for most graze planning work.   But, if  possible,  the  ob-
server  should  obtain  the  most  detailed topographic maps that
cover the selected area, so that observation sites can be  chosen
near  features  shown  on the maps.  Index maps are necessary for
ordering the topographic maps.  They may also give a  listing  of
local  map  dealers from which the maps can be purchased, and li-
braries that may have them.   Topographic maps are  discussed  in
more detail in Section 4.                                        
                                                                 
6.2.2.2  GRAZE PATH PLOTTING                                     
                                                                 
After obtaining the best map(s) for the location(s) selected, the
observer  plots the (sea level) limit line directly from the pre-
dictions.  To do this, the longitudes at the east and west  edges
of  the  topographic  map are found in the limit predictions, and
plotted at the  points (A and B in  Figure 6-4 in the  upper left
of the graphic shown here) where the limit
crosses  the edge meridians on the map (using the NORTH LATITUDE,
or 2nd and 3rd, columns of the predictions).  The line,  AB,  be-
tween  these two points is the sea level limit line.  Both points
will  often  not  be on the map and it may be necessary to tape a
piece of paper  to  the  map  to  extend  one  of  the  longitude
meridians.   Scales to facilitate plotting are given in "Appendix
C.  Plotting Scales" on  page 98 (not included here, but they can
be provided on request as  PostScript files that can be  read and
printed with Ghostview); scales for 1:250,000 and 1:125,000  maps
are in a Word document here.
                                                                 
If necessary, and always for elevations above 200 meters  or  500
feet,  the  limit line should be shifted to take into account the
elevation above sea level.    The  limit  lines  will  always  be
shifted south for high elevations in latitudes north of 31 deg. N 
(unless the elevation is below sea level).                          
                                                                 
MOON  AZIMUTH and TANZ in the graze limit predictions are used to
compute the corrections, as described in  "Elevation  Correction"
on  page  35, using Eqn. 6-1.  The sea level and corrected limits
are shown in Figures 6-3 and 6-4 in the upper part of the graphic 
shown here.  As  explained  in  "Elevation
Correction", if the elevation above sea level is  h              
(in  feet or meters), the magnitude of the correction to be ap-  
plied,  d, is given by the formula 
                              
                d = (TANZ) h   

where d is applied in the direction of the Moon's azimuth.       
The shift x is negative if the shift is toward the south for     
heights  above sea  level.                                       
                                                                 
An example of a corrected limit line is shown in Figure 6-4.  The
instructions for producing such a correction are as follows:     
                                                                 
1.  Using  a protactor, plot the Moon's azimuth (eighth column of
    the predictions) at point A; it is angle G or angle VAH,  al-
    ways measured clockwise from north.                          
                                                                 
2.  From the topographic chart, determine the approximate average
    elevation where observations will be made; this is h         
    (shown in Figure 6-3).  Multiply h                           
    by TANZ (ninth column in the predictions) to get d,          
    the distance the point A should be shifted in the direction  
    of the star's azimuth.  The distance AH is d                 
    in  Figure 6-4; therefore, since G is the Moon's azimuth, H  
    is a point in the elevation-corrected limit.  F or angle  VAI
    is  the azimuth (or bearing) of the limit and can be measured
    with a protractor; therefore D or angle IAH is the difference
    of G and F.                                                  
                                                                 
3.  The perpendicular shift of the limit, x                      
     or HI, to take elevation into account is then (AH) sin(D),  
    or (d) sin(D), or (h)(TANZ)sin(D).                           
                                                                 
4.  Now, draw the line JK parallel to the limit line  AB  through
    H; JK is the northern limit corrected for elevation above sea
    level.                                                       
                                                                 
In  the example, if sin (D) TANZ is given in the predictions, the
process is simplified, since the Moon's azimuth does not need  to
be plotted.  Just obtain x by multiplying h by sin(D) TANZ,      
and plot the point H a distance x                                
perpendicular to the sea level limit line, on the north side of  
the limit if sin (D) TANZ is positive and on the south side if it
is negative.                                                     
                                                                 
6.2.2.3  PROFILE USE                                             
                                                                 
The profiles are used to select the  region  of  the  graze  path
where  observers should be concentrated.  Once the graze path has
been plotted, the observers can be placed north and/or  south  of
the predicted limit where the tracks across the profile would in-
dicate  that  multiple  events  will  be seen.   For any location
within a few miles of the predicted limit the star will appear to
move in a horizontal line across the profile chart, as  shown  in
the example in Figure 6-5 in the  lower part of the graphic shown 
here.  The region of the profile chart where
two  or  more  mountains will occult the star is indicated by two
dashed horizontal lines on the profile, and is labeled "L".  This
is the suggested range.  It should be extended north and south by
the  amount  W (in seconds of arc, or left hand scale on the pro-
file), which is the probable error of the star's declination;  it
is given in the heading of the limit predictions.  This makes the
suggested  range NM in Figure 6-5.  The range in miles or kilome-
ters is determined by using the vertical scale on the right  side
of  the  profile.    With this scale, the suggested range NM, the
predicted limit line corrected for elevation, and the  perpendic-
ular  distances,  XM  and XN (Figure 6-4), from the limit line to
the edges of the range can be determined.    These  are  used  to
cunstruct  the  lines through points N and M that are parallel to
the corrected limit JK.  These lines are LP and QR, respectively.
The observing sites should be selected between these  two  lines.
In  the example in Figure 6-4, there are three sites along a road
within the range.                                                
                                                                 
Expeditions involving two or more stations should usually  try to 
spread  vertically across the path by about 1.0" 
of lunar graze height.  This is always more than a mile in dis-  
tance  perpendicular  to  the graze path.    But if the predicted 
range of  multiple  events  is very narrow, then a narrower range
can be covered, especially when the profile is based on  previous
graze observations (3's points).                                 
                                                                 
Once the sites have been selected, the next step is to  determine
the  time of the earliest event as seen from the sites.  The Uni-
versal Time of central graze is determined by  interpolating  the
times  given in the predictions between the longitudes of A and B
in Figure 6-4.  Near the top of Figure 6-5, the time with respect
to central graze is given in minutes, with negative values  being
times earlier than central graze time.                           
                                                                 
The  earliest event in the selected range on the profile chart is
where the dashed line marking the lower  edge  of  the  suggested
range intersects the lunar profile, usually before central graze.
In  Figure 6-5, it is at S, which is a disappearance about 1m 20s
before central graze.  So, observers would expect  to  see  their
first  event  about  the time of central graze minus 1m 20s.  Ob-
servers should be ready to time events about a minute or two  be-
fore  the  expected  time  of the first event.  Unexpected shifts
(prediction errors) can cause events to be early.                
                                                                 
Using the profile chart, it is hypothetically possible to predict
the sequence of events as seen from the three sites.   The  three
horizontal lines labeled 1, 2, and 3 in Figure 6-5 are plotted at
the  perpendicular  distances  of sites 1, 2, and 3 from the cor-
rected limit line in Figure 6-4.   Where these  horizontal  lines
intersect  the  profile  is when, relative to the time of central
graze, the events will occur.                                    
                                                                 
In the case of a known double star where there is  a  possibility
of  observing  the secondary, two lunar profiles will be provided
on the profile chart, one for each component.  Observers can  use
the  profiles to position themselves to observe grazing phenomena
of  either or both components.  In general, the expedition leader
should station most observers in the zone where  multiple  events
can  be  seen for the component whose profile is deepest into the
lunar shadow.  Then, most observers should see some complete dis-
appearances of the star, which are usually easier to observe than
the partial drop in light when one of the two  stars  disappears.
However, if the difference in component magnitude is 1.5m,       
it  is  best  to  position  observers  for the profile of the    
brighter component (indicated by P, for primary component).      
                                                                 
Often, the two stars will be so close that observers will not  be
able to tell that the star is double until the graze begins.  Due
to the grazing geometry, the step-wise events sometimes seen dur-
ing total occultations of double stars are much prolonged.  Visu-
ally  observed  grazes  in  flat  lunar  terrain are capable of a
resolution of 0.01",                                             
nearly equivalent to the resolutions obtained  in  photoelec-    
tric  work  with total occultations.   Several close double stars
have been discovered during grazing occultations.  However, grad-
ual or fading events seen during grazes are  more  often  due  to
grazing  enhancement  of  Fresnel  diffraction at the Moon's limb
rather than to stellar duplicity, which  is  more  noticeable  by
events occurring in distinct steps.                              
                                                                 
PROFILE CORRECTION FOR LONGITUDE:  Each profile is prepared for a
specific  position  angle  of central graze, as given on the pro-
file; it is automatically adjusted from the position angle at the
"standard" longitude to a point in the limit closest to  the  ob-
server's  station.  If observations will be made from a different
location (for example, if weather forces an  observer  to  travel
far  to  the east or west of the closest point), the profile will
require correction, which amounts to shifting the geometry of the
observer and lunar surface lines.   Motion in longitude  has  two
effects  on  grazing  occultation position angles:  a shift along
the limb, which is the dominant effect, and motion along the sur-
face, a much smaller effect near the lunar poles,  but  dependent
on the lunar motion and librations at graze time.                
                                                                 
There  are  two  methods for translating a profile to a different
longitude.  One, which will not be covered here, involves drawing
a new reduction profile using data from  the  USNO  OCC  program.
The  other  method  requires  drawing a new line of central graze
(CG) and a new predicted limit line on the existing  profile  for
the distant expedition.                                          
                                                                 
The  Watts  angle  (WA) of the new line of central graze (WAn) is
calculated by the equation given below; WA has the same scale and
direction as the position angle, PA, but is virtually always off-
set with respect to it.  Let us define the following quantities: 
                                                                 
PAO  the  position angle of CG for the original profile (given as
     ]POS ANGLE XXX.XXX PROFILE FOR ¼¼¼Ø in the footer data).    
                                                                 
PAN  the position angle of CG for the distant expedition (as  de-
     termined from the limit prediction).                        
                                                                 
WAO  the Watts angle of CG for the original profile.             
                                                                 
Then we have the relation:                                       
                                                                 
         'WA' sub n = 'WA' sub o + ('PA' sub n - 'PA' sub o)     
                                                                 
where WAn is corrected by 360 deg. if negative or -360 deg.      
                                                                 
Now  draw  a new line of CG for the distant expedition vertically
at an WA of WAn.  The new limit's intersection point with the new
CG line is determined by maintaining the limit line to mean  limb
distance  (the  HEIGHT  given  in the footer data) for both expe-
ditions.                                                         
                                                                 
Because of the vertical exaggeration (VE) of profile  plots,  the
new predicted limit line will slant with respect to the original,
especially if the two expeditions had a significant r difference.
The slope, s, of the new limit is determined as follows; let     
                                                                 
16.2766     the  number  of arcseconds that 1d of the Moon's limb
            subtends at its mean distance.  It is the product  of
            932.58,  the mean apparent angular radius of the Moon
            in seconds of arc, as subtended from the  Earth,  and
            0.01745329,  the  conversion  factor  from degrees to
            radians.                                             
                                                                 
R           the ratio of vertical distance on the profile  (miles
            or  kilometers) to horizontal distance for one degree
            (1þ) of WA (a measured quantity).                    
                                                                 
VPS         the vertical profile scale given in the footer data. 
                                                                 
Then the vertical exaggeration is:                               
                                                                 
                   'VE' = < 16.2766 R > over 'VPS'               
and                                                              
                         eqn. 6.4                                
              's' = tan (PA sub o - PA sub n) times 'VE'         
                                                                 
The time scale, or horizontal profile scale (HPS), will change if
the two locations are a large distance apart.  If they are within
several hundred miles of each other, the change should be  small.
If  no HPS is available for the distant expedition, a first-order
correction can be made.   The  equation  below  approximates  the
amount  that the Earth's rotation vector subtracts from the lunar
orbital vector, based on the lunar altitude  and  the  observer's
latitude;  the  calculation  is not valid over extreme distances.
Define:                                                          
                                                                 
AN      the  lunar  altitude of the distant expedition, as deter-
        mined from the limit prediction;  extrapolate  if  neces-
        sary.                                                    
                                                                 
AO      the lunar altitude of the original expedition.           
                                                                 
D       the  average of the geographical latitudes of the two ex-
        peditions.                                               
                                                                 
HPSO    the HPS of the original expedition.                      
                                                                 
HPSN    The HPS of the new expedition.                           
                                                                 
Then the new horizontal profile scale is:                        
                                                                 
 'HPS' sub n = 'HPS' sub o + 0.92 (sin a sub o - sin a sub n) cos
                                                                 
Note that HPSn is applied in the original  horizontal  direction,
not the slanted direction of the new limit.                      
                                                                 
                                                                 
7.0  ORGANIZING GRAZING OCCULTATION EXPEDITIONS                  
_______________________________________________                  
                                                                 
Grazing  occultations are most successfully observed when a group
of people meets at a prearranged site and sets up a chain of  ob-
servers  across  the  graze path.   Someone must obtain the graze
predictions, select the best site, notify  other  observers,  and
assign  observing  locations before the graze. Afterward, someone
must determine the location of each of the observing  sites,  re-
port the locations, and collect and report the observations.  The
graze  organizer  obviously  bears  the burden of the work.  This
section is intended to detail the steps involved in organizing  a
successful  graze.  A single observer, acting on his own, can ob-
serve a chord across the graze, and these data are useful,  espe-
cially  if  they  can  be  combined  with other cords obtained by
observers to the east or west during the same  graze.    However,
the  resolution  obtained by having a line of well-spaced observ-
ers, even if only a few, across the predicted graze path, is much
better, and worth the time it takes to set up such an event.  Ob-
server safety should be the top priority of all expedition  lead-
ers.    The  articles  discussing  expedition safety (Occultation
Vol., 5,  No. 11,  1993 March, pp. 284-286) should be read by all
graze expedition leaders.                          
                                                                 
7.1  PREPARATION                                                 
________________                                                 
                                                                 
The  time  and effort needed for organizing a graze will vary de-
pending on the number and experience of the  people  expected  to
observe.  As a general rule, the more people who want to observe,
the more elaborate the preparations must be.  The graze organizer
will  need  to select the sites, notify the observers, often help
the observers find equipment and transportation,  and  teach  the
new observers what to do.                                        
                                                                 
7.2  SITE SELECTION                                              
___________________                                              
                                                                 
The  graze  organizer  first  needs  to obtain predictions of the
grazing occultations for his local area.  These are available  to
members  of IOTA.  Also, predictions for the brightest events are
published in Sky and Telescope, the Handbook of the Royal  Astro-
nomical Society of Canada, etc., for the areas those publications
cover.  A graze organizer, however, will usually want the predic-
tions distributed by IOTA. These are the predictions discussed in
"Grazing Occultation Predictions" on page 29.                    
                                                                 
The  graze  predictions  can  be  used to determine approximately
where the grazes might be observed.  Any map of the state or gen-
eral region that gives latitude and longitude  can  be  used  for
predicting  where the grazes can be observed. Aeronautical charts
obtainable at airports, topographic index maps, atlas maps,  maps
from  National Geographic, and even the rare road maps with lati-
tude and longitude lines are often suitable.  The 1:250,000  maps
from  the  USGS and most other national mapping agencies can also
be used.                                                         
                                                                 
The predicted graze limit line at sea level can be plotted on the
maps directly from the predictions.  If the edges of the map cor-
respond to longitudes in the predictions, the organizer can  plot
the points where the limit crosses those edges by using the lati-
tudes  that accompany the edge longitudes.  The plotting of limit
lines and correcting them for elevation when necessary, are  dis-
cussed in "Profile Predictions" on page 35.                      
                                                                 
The  region  of  expected  events can be determined from the pre-
dicted profile, as discussed in "Grazing Occultation Predictions"
on page 29.  This region is the area  near  the  predicted  limit
where observers should be stationed.  It will be more than a mile
(1.6 km), and sometimes several miles (¼5 km) in length.  The ob-
serving  sites  should  be selected to fall within this suggested
range.  It should be stressed that the suggested  range  and  the
profile  are predictions of what is going to happen, and that in-
accuracies in the predicted stellar and lunar positions, as  well
as  the  lunar  profile, will mean that the exact location having
the most events cannot be precisely  predicted.    All  locations
within  the suggested observing range will have approximately the
same chance of seeing multiple  events,  with  the  edges  having
slightly less chance.                                            
                                                                 
Features to consider when selecting the sites from among the can-
didates are the ability to see the Moon at the time of graze from
the  site, the accessibility of the site to the observers, light-
ing, parking, traffic if the site is a road, what  permission  is
needed  if  the  site is private property, and whether public au-
thorities will need to be notified if the site  is  public  prop-
erty.   (It is advisable to notify local authorities, such as the
sheriff's office, or the office of the local police force,  espe-
cially  for expeditions with three or more sites.)  Also, the ob-
servers will need reference points to locate their stations  when
their positions are determined.                                  
                                                                 
Observers  will  want  to be able to move their equipment to find
the best observing locations for grazes at low Moon altitude.  It
may not be possible to determine the effects of local  obstacles,
such  as  houses, trees, and hills until actually arriving at the
graze location to observe.                                       
                                                                 
If the observing sites chosen are along a  road,  observers  with
cars will need places to park completely out of the traffic.  The
shoulder may be sufficient, but the smaller country lanes may not
have  them.  Also, if the road is heavily traveled by trucks, the
observers will need to be well away from  the  traffic  to  avoid
problems from equipment vibration and traffic noise.             
                                                                 
Railway tracks and railway rights of way are not public property;
they  are  the property of the railroad.  They should not be used
by graze expeditions without permission.  As a practical  matter,
even  if  the  observers will not be in the right of way but only
nearby, it might be wise to contact the rail line to determine if
there are any trains scheduled near the time of the graze.    The
vibrations  from  a train passing nearby can shake a telescope so
that observations are impossible.                                
                                                                 
Public parks may be closed after sunset.  If accessible, they are
often very good sites, because they do not have  many  lights  or
much traffic.                                                    
                                                                 
The organizer may want to inspect the sites before the graze, ei-
ther  to  check  their  suitability, or to mark the locations se-
lected for the sites.   If the  observing  location  selected  is
devoid  of  landmarks,  the observers are not going to find their
stations unless they are marked.  This can be done with marked or
numbered stakes pounded into the ground, cards tied (not  nailed,
that  is not always legal) to utility poles.  Water-soluble spray
paint or chalk can be used to write station numbers at  the  side
of a road.                                                       
                                                                 
7.2.1  OBSERVER NOTIFICATION AND PREPARATION                     
____________________________________________                     
                                                                 
Notification  of  the  observers  can be as simple as a telephone
call, or through a local astronomy club  newsletter,  or  through
graze  notices sent to anyone who has expressed an interest.  The
notification should tell the observers when and where  the  graze
will be, a meeting time and place, a telephone number to call for
notice of weather cancellation, and what equipment they will need
for  observing.  The graze organizer will want to know who is in-
terested in observing a particular graze, to help in planning the
observations.                                                    
                                                                 
7.3  EXPEDITION REPORTING                                        
_________________________                                        
                                                                 
The expedition leader needs to collect the observed times, or the
unreduced observation tapes, for all observers.  Also, the  accu-
rate geodetic coordinates for all successful stations must be de-
termined  (see  "Accurate  Topocentric Position Determination" on
page 18), and the expedition leader usually must take  charge  of
that  task  as well.   See "Appendix F. Report Forms" on page 106
for information on how to prepare the final expedition report.   
                                                                 
7.4  APPROXIMATE REDUCTION AND SHIFT DETERMINATION               
__________________________________________________               
                                                                 
The expedition leader should make an approximate reduction of the
observations by plotting the observed timings made  by  each  ob-
server  on  the predicted profile.  This allows an estimate to be
made of the shift of the graze shadow from its nominal,  or  pre-
dicted,  position.  The graze shift is the distance and direction
the predicted shadow of the Moon would need to be  moved  on  the
ground to match the actual shadow observed during the graze;  its
scale  is in seconds of arc subtended at the Moon's mean distance
and is the vertical scale on the right side of IOTA's ACLPPP pre-
dicted profiles.  The shift may be thought of as the residual  of
the  observations  for  the entire expedition; it is used to warn
future expedition organizers of stars with poorly determined  po-
sitions  and  to identify poor quality Watts limb correction data
(which are used to generate profile plots).    This  section  de-
scribes  the  procedures  used  to plot graze observations on the
predicted profile in order to determine this shift.  Plotting the
expedition's data is the best way to verify that all stations  of
a graze agree with one another.  Plots can also reveal large, un-
predicted  mountains  or valleys on the lunar limb that should be
added to the data set used in predicting profiles.               
                                                                 
All processes should be applied to every  successful  station  of
the  expedition, even though the process is described as singular
here.  If a value is needed from the limit prediction (the  sheet
that  gives the latitudes and longitudes of points in the limit),
it should be determined for the specific longitude of the  actual
expedition.   This discussion assumes that an IOTA Automatic Com-
                                              ____               
puter Lunar Profile Printing Program (ACLPPP)  profile  is  being
used; by convention north is at the top and time increases toward
the right for these plots.                                       
                                                                 
Plotting  the  observations on the predicted profile is very much
like the process of determining predicted event  times  described
in "Grazing Occultation Predictions" on page 29.                 
                                                                 
7.4.1  CORRECTION OF PROFILE TO ACTUAL OBSERVING LOCATION        
_________________________________________________________        
                                                                 
The  predicted profile is generated for a point in the limit near
the closest approach to the coordinates  the  observer  specified
when requesting IOTA predictions; this is usually one's home.  If
no  point  was specified, then an approximate location would have
been established by IOTA based on the mailing address  (latitude,
longitude, and elevation), along with very small (3-mile) default
travel radii for marginal, favorable, and spectacular grazes; one
should  contact  the IOTA secretary giving specific values if the
defaults are not acceptable.  If the graze was not observable  at
the  point  of closest approach due to a low lunar altitude or to
strong twilight or daylight there, then the profile will be  gen-
erated  for  the  first point in the limit at which the graze can
actually be observed.  A correction will usually be needed if the
actual location of the expedition was either  up  or  down  track
from the point for which the profile was generated.  The position
angle  of graze for the actual location of the expedition, PA(a),
is determined from the limit prediction (see Figure  7-1).    The
position  angle  for  which  the profile was generated, PA(p), is
given in the footer of the profile as "POS ANGLE XXX.XXX  PROFILE
FOR (your name)" (see Figure 7-2).  If PA(a) and PA(p) are within
approximately 0.2 deg.,                                          
the correction is small and may be neglected.                    
For a difference >0.2 deg.,                                      
a new line of central graze and a new sloping predicted limit    
line  should  be drawn on the profile for the actual location.(7)
It is necessary to use the Watts angle scale to position the  new
line of central graze, since position angle is not graphed on the
profile;  the two have the same scale and direction, but are usu-
ally shifted with respect to one another.  Let WA(p) be the  cen-
tral  graze  Watts angle for which the profile was plotted (given
in the footer data); then the Watts angle of the new line of cen-
tral graze, WA(a), is given by  the  equation  (similar  to  Eqn.
6.2):                                                            
                 WA(a) = WA(p) + lb PA(a) - PA(p) rb             
                                                                 
If  WA(a)  does not fall within the range of 0 to 360d, it should
be normalized by either adding or subtracting 360.  A new line of
central graze for the actual location is then drawn vertically at
a Watts angle of WA(a).  The new limit's intersection point  with
the new central graze line is at the same distance above the mean
limb  as  the original limit was above the mean limb.  Due to the
vertical exaggeration (VE) of profile plots,  the  new  predicted
limit  line  will  slope  with  respect to the original profile's
limit.  Slope is defined as the change in the vertical (y)       
coordinate divided by the change in the horizontal (x)           
coordinate.  The vertical exaggeration and slope are  then  de-  
termined  as  described  in "Profile Correction for Longitude" on
page 44.                                                         
                                                                 
When using this method to combine the results of two or more  ex-
peditions  on  one profile, the profile of one of the expeditions
should be considered the "original" profile (the  one  calculated
for  the  "point closest to home" in the previous discussion) and
the other expedition's profile should be considered the "actual",
or "new" location.  More than two expeditions may be combined  by
7   This procedure can also be used to  combine  the  results  of
separate graze expeditions on one predicted profile.             
                                                                 
having  multiple "actual" locations.  If the HEIGHT values listed
in the footer data of the profiles differ significantly, the dif-
ference should be applied when plotting the actual limit's verti-
cal  position.    (These  HEIGHT values represent a set of global
empirical vertical corrections applied to the lunar mean limb  at
the  time  the  profile is generated).   The same is true for the
HPS; since its value is available for both locations when combin-
ing results, any difference between expeditions  should  also  be
taken into account when plotting the observations.               
                                                                 
A fictitious example using an actual graze prediction follows (no
expedition was actually attempted).  Figure 7-2 is a profile gen-
erated for a position angle PA(p) of 345.38                      
(POS  ANGLE 345.380 PROFILE FOR DON STOCKBAUER).  If an expe-    
dition had actually obtained data at longitude 99.75 deg.        
West, the actual position angle of graze, PA(a), would be 344.7d 
(from the limit prediction, Figure 7-1).  The Watts  angle  for  
which  the profile was generated, WA(p), is 5.29; the Watts angle
at which to draw the new line of central graze for the actual lo-
cation, WA(a), is  5.29 + 344.7 - 345.38 = 4.61 deg.             
Since the predicted limit was scaled  off  the  top  of  the     
northern  boundary  of the plot, it is drawn in as a dotted line.
The original limit is 0.77 miles north of the mean limb  (repres-
ented  by  the  series of D's), so the actual limit must maintain
this distance above the mean limb at its new position.  The ratio
of one  mile  to  one  degree  of  Watts  angle  is  measured  as
(1.04/0.74) inch, which equals 1.40.  The VPS is given as 0.85"  
per mile in  the profile's footer data.  The vertical exagger-   
ation is (see Eqn. 6.3):                                         
            VE = ( 16.2766 * 1.40 ) over 0.85  = 26.81,          
and the new limit's slope (from Eqn. 6.4) is:                    
          s = tan % ( 344.7 - 345.38 ) * 26.81 = -0.318.         
                                                                 
The new limit for the actual  location  slopes  0.318  unit  down
(south)  for every unit of distance to the right (toward increas-
ing time).  "Graze height" is defined as an observer's perpendic-
ular distance on the ground  to  the  elevation-corrected  limit,
with  north  positive  by convention (not to be confused with the
HEIGHT shown in the profile's footer mentioned earlier).  In Fig-
ure 7-3, the graze height for an observer at point N would be NF;
since N is south of the corrected limit, it is a  negative  quan-
tity.    In Figure 7-2, parallel lines representing graze heights
of ¯1 to ¯4 miles are also drawn.  These graze heights are offset
vertically on the profile from the new sloping limit,  not  at  a
right  angle to it.  The graze height scales along the right-hand
border of the plot (in miles or  kilometers)  and  the  left-hand
border  (in seconds of arc) should be relabeled with the new val-
ues; the new values are circled in the example.                  
                                                                 
7.4.2  CORRECTING THE SEA LEVEL LIMIT PREDICTION FOR ELEVATION   
______________________________________________________________   
                                                                 
The limit prediction is calculated for sea level and needs a cor-
rection applied for other elevations.  No correction is needed if
all stations were less than 300 feet above sea level.  If all ob-
servers were situated at approximately  the  same  elevation  (to
within 150 feet), then an average correction based on their aver-
age  elevation may be used.  The elevation correction (EC) is de-
rived as follows:   a vector is constructed  originating  at  the
elevation  of  the  observer's location directed either toward or
away from the star, so positioned that it terminates on  the  sea
level  limit  (constructing  the  vector in this manner will give
north as a positive correction, and  south  as  negative).    The
product of the elevation and the tangent of the star's zenith an-
gle  (TANZ) gives the magnitude of this vector projected onto the
horizontal plane.  The sine of the difference between the limit's
bearing and the lunar azimuth (SIND) times the magnitude  of  the
vector  in  the  horizontal  plane gives the elevation correction
perpendicular to the sea level limit.   The elevation  correction
is given by Eqn. 6.1.                                            
                                                                 
If  the  limit prediction gives the product SIND * TANZ, then the
factor SIND need not be computed.   (TANZ is always  provided  in
the limit prediction either by itself or as a factor of the prod-
uct SIND * TANZ.)                                                
                                                                 
To calculate SIND, the Moon's azimuth for the location of the ac-
tual  expedition is determined from the limit prediction.  If the
prediction does not give the bearing of the limit, then  it  must
be  measured  using the plot of the limit on the topographic map.
It has values between 0þ and 180þ and is measured  eastward  from
north  (e.g., north = 0, east = 90, and south = 180þ).  Then SIND
= SIN (limit's bearing - Moon's  azimuth),  paying  attention  to
signs.                                                           
                                                                 
The  corrected  limit  should  be  drawn perpendicular to the sea
level limit at a distance equal to the magnitude of the elevation
correction; its direction is north of the sea level limit if  the
correction  was  positive,  or  south  if negative.   "north" and
"south" here only mean on which side of the sea level  limit  the
corrected limit should be drawn; the correction is always applied
at  a  right angle to the sea level limit, not due north or south
(unless, of course, the sea level limit has a  bearing  of  90þ).
As a check, the corrected limit should always be in the direction
of  the Moon's azimuth from the sea level limit for positive ele-
vations, and opposite it for negative  elevations  (this  can  be
checked  by plotting the azimuth of the Moon beside the predicted
limit on the map).   The  elevation  correction's  magnitude  can
range  from  zero to several times the observer's elevation.  For
locations at positive elevations north of latitude 31þ North, the
corrected limit will always lie south of  the  sea  level  limit.
The graze height is then measured for each station.              
                                                                 
Figure  7-3  represents a section of the topographic map used for
the fictitious graze in the previous example.    AB  is  the  sea
level  limit,  angle VAB is the limit's bearing, angle WBH is the
Moon's azimuth, HB is the elevation correction's projection  onto
the  horizontal  plane,  and HE (the elevation correction) is the
correction's projection perpendicular to  the  sea  level  limit.
The  elevation  of the observer is 2300 feet; TANZ from the limit
prediction (Figure 7-1) is 3.67, the sea level limit's bearing is
measured from the map as 105.8þ, and the lunar azimuth  from  the
limit prediction is 282.5 deg.                                   
The elevation correction HE is:                                  
sin % ( 105.8 - 282.5 )%%degrees * 3.67 * 2300%%feet = -486%%feet
                                                                 
The  corrected  limit is drawn as line CD, 486 feet perpendicular
to the sea level limit; since the correction is negative,  it  is
drawn on the south side of AB.                                   
                                                                 
7.4.3  CALCULATING THE TIME OF CENTRAL GRAZE                     
____________________________________________                     
                                                                 
The  Coordinated  Universal  Time (UTC) when the line of sight to
the star passed closest to the apparent center of the Moon's disk
is then calculated for each observer; this is called the time  of
central graze, or just central graze.  If one uses the method de-
scribed  in "Calculating the Shift Value" on page 55 to determine
the shift (that of fitting the profile to the observations by al-
lowing both horizontal and vertical movement), then the following
simplified method will suffice to determine central  graze.    In
Figure  7-3, central graze for a site at N would be determined by
interpolating  for  the  time  at  point  R  (N's   perpendicular
projection onto the sea level limit) using the times given in the
predictions  for  points  A and B.  If a precise absolute time of
central graze is needed, then station N's location would need  to
be  projected  onto  the sea level limit using the Moon's azimuth
instead of a perpendicular projection; that point would  then  be
used  for the interpolation or extrapolation.  This is due to the
fact that curves forming equal times of central  graze  propagate
in the direction of the Moon's azimuth, which is not usually at a
right  angle  to  the limit.   However, the simplified method has
traditionally been used because if the lunar altitude is low  (as
in  this  example), the limit's bearing and the lunar azimuth be-
come nearly equal.  The projection distance can then become quite
large; the data fitting  method  described  in  "Calculating  the
Shift  Value"  on  page  55 avoids this problem.   If all central
graze times for all stations of the expedition fall within a  one
second  time interval, then an average time may be used to repre-
sent them all.                                                   
                                                                 
In Figure 7-3, point R is 0.139 of the distance from point  A  to
B.   The time at point A is 5:23:45.6 UTC (determined by interpo-
lating for longitude 99 deg. 47' between 99 deg. 52.5'           
and  99 deg. 45'; point  A  is not at a standard topographic map 
boundary); the time at B is 5:23:46.6.    Interpolating  for  the
time at point R gives 5:23:45.7.                                 
                                                                 
7.4.4  PLOTTING THE OBSERVATIONS ON THE PREDICTED PROFILE        
_________________________________________________________        
                                                                 
A  line  is  then plotted that corresponds to the station's graze
height using the mile or kilometer scale on the  right-hand  side
of  the  profile,  or the slanting graze height coordinate system
for an adjusted profile.  This line represents the track  of  the
line  of sight from the observer to the star made across the pre-
dicted profile.   The differences between the  observed  time  of
each event and the central graze time are computed in minutes and
fractions  of  a minute.  The station's timings are plotted along
this line using these time differences; the scale used is the one
of minutes from central graze given at the top  of  the  profile.
Negative time differences are before central graze (to the left);
positive times are to the right.                                 
                                                                 
In  Figure  7-2,  the observer was stationed at a graze height of
-1.43 miles, or -1.22" (i.e., 1.43 miles, or 1.22" south  of  the
elevation corrected limit).  The observer had a disappearance (D)
at 5:22:17.7 UTC and a reappearance (R) at 5:24:55.7, which gives
time  differences  of  -1.467  minute  and +1.167 minute, respec-
tively.  The time differences are marked on the observer's  graze
height  line  by  measuring  horizontally to the adjusted line of
central graze, which is at Watts angle WA(a).   A  break  in  the
line  is  used  to  show  the period when the star was behind the
Moon.                                                            
                                                                 
7.4.5  CALCULATING THE SHIFT VALUE                               
__________________________________                               
                                                                 
                                                                 
In Figure 7-2, the scale on the left hand  side  of  the  profile
(the  unlabeled numbers) is in arc seconds and is the one used to
measure the shift.  Note that because this profile was  adjusted,
the  arcsecond  scale was also adjusted and is now represented by
the circled numbers.  The period when the star was observed to be
behind the Moon is higher (more  northerly)  than  the  predicted
profile would indicate, so a north shift was seen (i.e., the pre-
dicted profile would have to be shifted north to match the obser-
vations).  To measure the shift accurately, the predicted profile
information  is transferred to a translucent medium such as trac-
ing paper or a clear transparency.  The tracing is moved horizon-
tally and vertically with no diagonal skewing with respect to the
original until the predicted profile on the tracing fits the  ac-
tual  observations  best.   This takes liberties with the time of
central graze by making it a free parameter, but a better fit  is
almost  always  obtained  using this procedure because of various
inaccuracies present in the profile generation process.  However,
if the observations are shifted by a large extent with respect to
the calculated time of central graze (by about 20s or more), then
either a mistake was made or the prediction was grossly in error;
this  should  be  noted  when making an observation report.   The
shift (in tenths of an arcsecond) can then be  measured  directly
from  the  amount  the overlay had to be moved.  Its direction is
north if the overlay was moved upward, or south  if  moved  down-
ward.    Both the magnitude of the shift and its direction should
be reported.  For this example, an accurate  measurement  of  the
shift gives 0.5" north.                                          
                                                                 
The  shift  can also be calculated numerically.  If this is done,
an average should be taken based on  individual  values  measured
from 0.2 deg.                                                    
bins of Watts angle, not on individual values measured for each  
event timing.                                                    
                                                                 
From  time to time it is suggested in Occultation Newsletter that
an empirical correction be applied to a certain class of profiles
to better position graze stations based on trends  in  the  graze
shifts noticed by recent expeditions.  These suggested shifts are
for positioning observers only and should be ignored when comput-
ing and reporting the shifts for individual expeditions.         
                                                                 
IOTA's  grazing  occultation  coordinator can supply report forms
and instructions on their usage upon request, and can assist  ob-
servers with any aspect of the topics covered in this section.   
______________________________________________________________

Click here to obtain the prediction files for 
grazes in the northeastern U.S.A. (within 300 miles of Greenbelt, 
Maryland) for 2003.  Unzipping the file will produce 5 prediction files:

areg03.153   Grazing occultation limit predictions from Grazereg
argn03.153   ACLPPP profiles for the grazes in areg03.153
srsm03.153   The summary list at the top of areg03.153
alcm03.153   Asteroidal/planetary appulse local circumstance 
              predictions for Greenbelt, Maryland
aoccn153.003 Predictions of total lunar occultations for Greenbelt, 
              Maryland produced with Occult

Click here for plotting scales (in minutes of arc of 
latitude and decimal fractions) for the DeLorme Maryland/Delaware 
and Virginia Atlas and Gazetteers.