Return to my Astronomy/Space pages
Go to my home page

Let's Predict an Eclipse

© Copyright 1998, Jim Loy

Let's predict an eclipse (we see one on the right). This task may sound difficult, and I guess it is. But, most of the ideas are fairly simple.

The diagram, at the left, shows two eclipses. We see an eclipse of the sun on the right. In such an eclipse, the observer on the earth stands in the moon's shadow. We see an eclipse of the moon on the left. In this case, the moon is in the earth's shadow, and any observer on the night side of the earth can see the eclipse. So, we may deduce that seeing an eclipse of the sun is fairly rare. We have to be at the right place, on the earth's surface, at the right time. And seeing an eclipse of the moon should happen fairly often.

An eclipse of the sun occurs during a new moon. A new moon happens once a month. But we don't see an eclipse of the sun every month. The reason is that the moon passes to the north of the sun, or to the south of the sun, during most new moons (as in the right side of the second diagram). The same thing happens with eclipses of the moon. The moon passes to the north or south of the earth's shadow (the left side of the same diagram). And the reason for all this is that the plane of the moon's orbit around the earth is not the same as the plane of the earth's orbit around the sun. The moon's orbit is tilted about 5 degrees.

It is difficult to observe a new moon, as the moon is in shadow, and appears fairly close to the sun (either north or south of the sun). But, if we do observe several new moons, we may see the new moon a ways south of the sun. Then the next month the new moon will appear closer to the sun. Eventually, half a year after our first observation, the moon will appear a ways north of the sun. And then it will start to appear closer and closer to the sun, again. There is a point, twice a year, where the moon appears neither north nor south of the sun. Then we have an eclipse. These two points in the moon's orbit are called the ascending node, and the descending node. The ascending node happens when the moon is moving from south to north. And the descending node is when the moon is moving from north to south.

The moon passes through each node once a month. But, the sun passes through each node only once a year. So, a new moon or a full moon does not normally occur near a node. There is at least one eclipse, every new moon or full moon, when the sun is near one of these two nodes. Some of these eclipses may be partial, when the shadow of the earth is not centered on the moon, for example. Or, you the observer may be in the wrong place to see the eclipse. So, you do not see an eclipse every time. But, there are approximately four eclipses, every year.

The sun passed through the moon's ascending node on March 1, 1998. There was a total eclipse of the sun on Feb. 26. And there were eclipses approximately six months later. The sun passed through the moon's descending node on August 24, 1998. And an annular eclipse (not total, because the moon was too far away for its shadow to touch the earth) of the sun happened on Aug. 22. So, do eclipses normally happen in February and August? No. The ascending, and descending nodes move. They are rotating. The sun passes through these nodes about 19.6 days earlier, every year. That means that the sun will pass through the moon's ascending node earlier in February, in 1999, about Feb. 10. In order to find out about eclipses at that time, we have to figure out where the moon will be, at about that time.

New moons (and full moons) happen about every 29.53 days. Let's look at a table (beginning Jan. 12, 1998). The times are GMT:

The right column shows the time since the previous new moon. The next to the right column shows the time since the previous full moon. As you can see, these times oscillate. We have already predicted that the next ascending node will be about Feb. 7, 1999. We can predict, roughly, that we will have a full moon on Jan. 2, 1999, a new moon on Jan. 17, full moon Jan. 31, and new moon Feb. 16.

So, we have a full moon ten days before the sun passes through the moon's ascending node, and a new moon six days after the sun passes through the same ascending node. By comparing with similar situations, over the years, we can guess that there will be two partial eclipses, at this time. Jan. 31, 1999 will be a partial eclipse of the moon. And Feb. 16, 1999 will be a partial eclipse of the sun.

The sun passes through the moon's descending node near the beginning of August, 1999 (about Aug. 4). And, I read in the World Almanac that there will be a total eclipse of the sun on Aug. 11, 1999.

My predictions, above, are relatively inaccurate. As, I don't know exactly when new moons, and full moons will occur, in 1999. And, I don't know when the sun passed through the moon's descending node in 1998. If it was 1/2 year after the ascending node, then it was Aug. 18, 1998. Anyway, I predict two fairly uninteresting eclipses near the beginning of 1999.

If we knew the exact time of a new moon, we could tell which part of the earth will be facing the sun at that time. If this new moon is days after an ascending node, then the part of the earth experiencing the total eclipse would be north of the equator (as the moon is ascending northward). If it is days before the ascending node, then the eclipse would be south of the equator. The situation is the opposite for the descending node. You can also consider whether the earth is tilted toward or away from the moon. So, we can begin to see how we can pinpoint the location of a total solar eclipse.

With lunar eclipses, given the time of the full moon, we can tell which side of the earth is experiencing night, during the eclipse. So, we can deduce which parts of the earth can see the eclipse.

Note: Above, I said that the ascending node probably happened on Feb. 26, 1998. I deduced that because the eclipse on that day was seen almost exactly on the equator. But, the northern hemisphere was tilted away (a little) from the sun on that day, so the eclipse was north of the center of the earth. So, Feb. 26 was probably after the ascending node. I would like to look that up somewhere. I haven't found it, yet.

Warning: A solar eclipse can be dangerous to look at. You can be blinded by an eclipse. Even though the eclipse looks dark, there is plenty of light. And the pupil of your eye will be somewhat dilated, as the ultraviolet (and infrared, I am told) light will not be noticed by the retina of your eye. An eclipse can cause permanent damage to your eye. Watch it through welder's glass or something of that sort. Or, observe the image of it, projected by a lense or telescope, onto a white object. Never look at the sun through a telescope, without extreme precautions (partially blocked front of the telescope, and a dark filter). I have seen partial eclipses that way; it looks pretty good. I was able to see sunspots that way. I guess you cannot get a good view of the corona, that way.

Photographing the sun through a telescope can burn a hole through the camera's shutter! Think what it can do to your eye.

Update: I have made a few corrections in the above article.

I found a book with details on future eclipses. It seems that I did fairly well. The next two eclipses will be on Jan. 31, 1999 (lunar) and Feb. 16, 1999 (solar), as I predicted. But, the one on Feb. 16 will be an annular eclipse, not a partial eclipse.

I will work on this article some more. We will see if I can predict some more eclipses.

Of course, with more accurate data, I also have found very detailed descriptions of each future eclipse. It is a little tempting to predict eclipses for years to come, fudging the data to make the eclipses come out on the right days. I have avoided that, in the article above. My prediction of when the sun is at the moon's descending node in Aug. 1999 is a day (actually only a few hours) off.

Addendum:

An eclipse is an example of a syzygy (a favorite word in Scrabble), an aligning of planets or moons.

Above, I forgot to mention solar time. The sun may be up to 16 minutes of time from where the clock says it is. See Solar Time. This does not usually affect the date of an eclipse. But I should have mentioned it. It certainly affects the path of the moon's shadow on the earth.

Return to my Astronomy/Space pages
Go to my home page