Wednesday, April 23, 2014

Eclipses and Saros Cycles 3

This is the third part of a three part post. To begin viewing the first, click here.

The previous post revealed the primary astronomical tool for predicting eclipses, the saros cycle. According to this cycle, the Moon has the same apparent size, inclination to the Earth's orbit, and angle to the Earth-Sun line in the Earth's orbital plane every 6585.32 days. Therefore, if an eclipse occurs on a given date, another very similar eclipse will occur about 18 years later with respect to the above parameters. During an eclipse, the angle of the Earth-Sun line to the Earth-Moon line will either be 0° (for a solar eclipse) or 180° (for a lunar eclipse), and the Moon will be at the same angle after a saros cycle, so two eclipses separated by a saros cycle must either both be solar or both be lunar. Similarly, during an eclipse, the Moon is either passing through the Earth's orbital plane at the ascending node or the descending node, and the eclipse after a saros cycle has passed will involve the Moon at the same node. Finally, the apparent size of the Moon will be the same for eclipses separated by a saros cycle.

However, many other conditions are not invariant before and after the length of time of a saros cycle. For example, since the decimal part of the length of the saros cycle is .32 days, two eclipses separated by a saros cycle will not be at the same time of day on Earth. In fact, the second eclipse will occur about 8 hours later than the first (.32 days is approximately 8 hours). This also guarantees that the eclipse will not occur in the same place on the Earth's surface. Futhermore, since the saros cycle is about 18 years and 11 days, the Earth will also be in two different places on its orbit around the Sun when the two eclipses occur.

In addition, 6585.32 days is not exactly an integral multiple of each of the synodic, draconic, and anomalistic months. For instance, 242 draconic months is about 6585.35 days (using the more precise value of 27.21222 days for the draconic month), which, although very close to the length of 223 synodic months, 6585.32 days, is not equal to it. As a result, even as as 223 synodic months pass from one eclipse and the Moon returns to its position along the Earth-Sun axis, the period of 242 draconic months is not quite complete, so the Moon will not have quite reached its node. Thus, over time, the eclipses separated by saros cycles shift until, eventually, the necessary conditions no longer align, and eclipses cease. Any given saros series will therefore only have a certain number of eclipses before the alignment ends.



The above image illustrates the progress of a saros series of lunar eclipses, where the gray bullseye represents the Earth's shadow (recall that if the Moon passes through the light gray, or partial shadow, of the Earth, the eclipse is partial, and if it passes entirely into the dark gray area, the total shadow, the eclipse is total). This series of eclipses is also at the descending node. Therefore, we can see that, before the saros series begins, the Moon has already passed the descending node at the time of the correct angular alignment of the Moon, Earth, and Sun. Thus the Moon is "too low" to be hit by Earth's shadow (blue-gray Moon). However, after a saros cycle has passed, the angular alignment is once again correct, but a full draconic month has not yet passed, so the Moon is a little less past its descending node than before.

Eventually, the draconic month has drifted backward enough so that the passage through the descending node does coincide with the correct arrangement of the Moon, Earth, and Sun in the orbital plane, and the path of the Moon does take it through the Earth's shadow. At this point, partial and then total eclipses will occur. However, over time, the Moon's progress in the draconic month will again fall behind, and at the time of correct angular orientation, the Moon will have not yet reached the descending node. At this point (the pink Moon), eclipses will cease and the saros series ends.



A similar phenomenon occurs with solar eclipses. In the above diagram, a few total solar eclipse paths across the Earth are shown, all from the same saros series (though there are many others), and again at the Moon's descending node. Note that each eclipse is indicated by a path along the Earth's surface. The paths are the areas which experience totality some time during the eclipse. The path of the eclipse moves from west to east, even though the part of the Earth facing the Sun moves from east to west along the globe. This is because the Moon's velocity, and thus the velocity of its shadow, is greater from west to east then the area that of the point on the Earth's surface moving from east to west.

As mentioned above, successive eclipses in a saros cycle are separated (in time of day) by about 8 hours, approximately a third of a day. Thus the second eclipse takes place about 120° west of the first. Thus two eclipses in a saros series three saros cycles apart will be at about the same longitude. But for the above series, which take place at the Moon's descending node, eclipses later in the series are not as far along with respect to the draconic month as those at the beginning. Thus, just as for lunar eclipses, the eclipses in a series at the descending node shift "upward", and in the solar case, this has the effect of shifting the totality path north for each successive saros cycle. Thus the first eclipses in the same saros series as above, are only partial, and can only be seen near the south pole. Eventually, as the series progresses, the Moon's shadow intersects the Earth, again beginning at low latitudes. As in the figure, the path shifts north until the eclipses can only be seen near the north pole, and then not at all. The down-to-up or south-to-north pattern for saros series reverses for series at the ascending node.

Saros series are numbered, with separate numberings for lunar and solar eclipse series. For lunar eclipses, those at the descending node are given odd numbers and those at the ascending node even numbers. The situation is reversed for solar eclipses. The saros series, further, are ordered in number by starting date. Contemporary series which are currently producing total eclipses are in the low to mid 100's. For example, the solar eclipse on August 21, 2017 belongs to solar saros 145. Solar and lunar saros series each include around 70 eclipse events, spanning a period of roughly 1200 years. Of course, these series overlap, so any given year, eclipses occur from different saros series. For example, the year 2012 had solar eclipses from series 128 and 133, while it had lunar eclipses from series 140 and 145.

In conclusion, saros series are the most important tool in the categorization of eclipses, and have allowed us to identify when eclipses have occurred thousands of years into the past, and to predict when the will occur thousands of years into the future.

Sources: http://en.wikipedia.org/wiki/Saros_(astronomy), http://www.cropcircleconnector.com/inter2012/italy/bracciano-earthquake7.jpg, http://www.math.nus.edu.sg/aslaksen/gem-projects/hm/0304-1-08-eclipse/Path%20of%20Totality.htm, http://eclipse.gsfc.nasa.gov/SEsaros/SEperiodicity.html#section106

No comments: