Mr. Dennette (Abu-Hurairah) Harrod, Jr
Dennette@WiZ-WORX.com
URL: http://www.wiz-worx.com/
December 27, 1995
From: Abu Hurairah Ibn Dennette (dennette@WiZ-WORX.com)
To: editor.ar-risalah@ISGL.org (Ar-Risalah Newsletter: Islamic
Society of Greater Lowell)
Subj: Halal sightings for Ramadan and Eid
Date: 27-Dec-95
Bismillah-ir-Rahman-ir-Raheem ... As-salaamu-alaykum, my brothers and sisters in Islam.
Insha-allah, we shall soon begin the holy month of Ramadan and fulfill one of our sacred obligations as True Believers. As a muslim scientist, I participated in an international conference on the lunar calendar in 1988. Using algorithms presented at that conference, I have written a computer program to display halal (lunar crescent visibility) for sundown on the day of any New Moon.
The chart above is for the first crescent of this month, which occurred a few days ago. Any community which has a halal committee would have attempted to observe that crescent, if only to verify the suitability of the viewing site for the start and end of Ramadan.
The large (green) *s are places from which the crescent should be visible to an unassisted observer at ground level (i.e., no infrared binoculars from an airplane above the clouds). The smaller (red) *s represent a "zone of uncertainty" in which an observer, even under ideal conditions, may or may not be able to witness the astronomically defined event. The areas with no dots will not be able to see a crescent at sundown on that date.
The "trash" at the bottom of the world? A mathematical anomaly, like attempting to take the square root of a negative number. In this case, it occurs in the polar regions at the times of the midnight sun. Coincidentally, this lunation coincides with the shortest day of the year in the Northern Hemisphere (where we live) and the astronomical first day of winter, which is the first day of summer in the Southern Hemisphere.
The criteria for visibility used by the program are simple: when the sun is so-many degrees below the horizon, is the moon high enough above the horizon to be seen by the unaided human eye? If it is, then it gets a big (green) *. If not, does it still fall within a range of tolerance; if so, then it gets a small (red) *. Otherwise, it cannot be seen, so there is just as faint dot that shows the location has been examined.
This test is performed for every location on a grid that is five degrees of latitude and longitude on each axis, from +70 to -70 degrees latitude ... over ten thousand locations. The "zone of uncertainty" corresponds to the "range of tolerance" in the algorithm. This is to make allowance for such factors as
As we all know, a half-moon can be see either setting in the East just before noon, or rising in the East just after noon. At this latitude, the sun does not get far enough below the horizon for the moon to be visible, but the algorithm "ignores" that fact, simply verifying that the angular distance between the sun and moon is sufficiently large that if the sun were far enough below the horizon, then it would be visible.
There is another reason for the "zone of uncertainty" which has to do with the amount of effort put into the calculations. Some short-cuts have been taken in the implementation of the algorithm because the number of variables that must be considered increases both the size and execution time of the program, and some cannot be known except by direct observation (i.e, what are the temperature and barometric pressure?) at the time and location for which the calculations are being made. This kind of defeats the purpose of making the calculations. (If you're there to take the temperature, you can look West to see if there is a crescent moon in the sky.) So we assume a constant (average) value for all locations, just as we have assumed a constant diameter of the earth that we use for the "sea level" altitude value. (Are you beginning to get the picture?)
Even if someone were willing to sift through the satellite data and provide a data base with altitudes for all of the locations for which calculations are made, it's kind of academic because "when the sun is so-many degrees below the horizon" is a time that has to be either calculated for each location, or assumed to be a constant.
The problem is that although it is possible to calculate the moment at which the sun's disk is bisected by the horizon (the angular position of its geometric center is zero degrees), that calculation involves things like altitude (actually, distance from the center of the earth) and atmospheric refraction (temperature, humidity, pressure are all assumed constant, which also speeds things up a bit), it requires an additional iteration of the algorithm with new coefficients based on the angle of inclination of the sun's orbit, which changes from day to day ... it's steeper in the Winter and shallower in the Summer, which is another reason why the time between sundown and the end of twilight is shorter in the winter and longer in the summer.
Thus, "when the sun is so-many degrees ..." is different for each location for every day of the year. Sometimes it's ten minutes after sundown, some times it takes twenty minutes. Adding that calculation also adds time to the execution of the program, and there are 10,152 points on the grid! So, the choice is to get the calculations in less time than it takes to get a cup of coffee, or set the program to run overnight while you sleep and look at the results in the morning ... but you're still "uncertain" because of the atmospheric conditions, like air pollution!
To get back to the crescent moon visibility of 22-Dec-95, this spherical projection of the earth centered above New England (actually, the Drum Hill Rotary in North Chelmsford, MA) shows that everyone in North America (except for northern Canada and most of Greenland) should have been able to see the halal at sundown on that date, assuming the sky was not overcast and that they had an unobstructed view of the horizon.
What is not so cut and dry is the situation in Asia, because the "zone of uncertainty" passes right through the Middle East. It is irrefutable that muslims in Indonesia, Malaysia, and the Philippines would not be able to see the halal at sunset, but in Mecca and Medina? Well ... Allah knows best if they saw it or not. If it's worth a long distance phone call and you know someone over there that you can ask, then maybe you can find out.
This is where the rubber meets the road ... the halal that signals the start of Ramadan. All of the calendars say that the New Moon occurs on 20-Jan-96. All of the muslim astronomers are of the opinion that it will be visible in Hawaii, that it is "questionable" if it will be visible on the western coast of North America, and that it will be impossible to see on the eastern coast of North America or anywhere else in the world.
It is not my place to say that the Ummah should not begin their fast of Ramadan on Sunday. In my own heart, I feel that I should begin my fast on Monday, 22-Jan-96. The ISNA fiqh Committee may decide that Hawaii is a part of the United States, so that all Muslims in North America should begin their fast on Sunday, 21-Jan-96.
The real problem is the Eid, because the "zone of uncertainty" once again passes through the Middle East, and the question is once again raised, "Why do we celebrate the Eid on one day but they do not do it until the next day in Singapore", while in Malaysia the people ask, "Why is my neighbor ending his fast a day before my family just because his relatives in Cairo are all ending their fast the day before we are?"
It is not my desire to add to the confusion around these questions, but to fulfill my obligation as a muslim to educate my brothers and sisters as to the reasons why science alone cannot provide a "definitive" answer to this question that gets raise and argued every year. I don't make the rules, I just explain the laws that define them and I try to obey them as best as I can interpret them, with His help.
That is all I have to say.
DAHarrod