How come the Moon isn't invisible during a total lunar eclipse? And what is the Danjon scale? [Answer]
What are the characteristics of the solar eclipse track of annularity and/or totality? In which direction does the solar eclipse track trace on Earth? Are there exceptions? [Answer]
During a solar eclipse, how fast does the lunar umbra moves on Earth? [Answer]
Is it that only partial solar eclipses are visible from the poles? [Answer]
What are the lines "eclipse begins/ends at sunrise/sunset" and "maximum eclipse at sunrise/sunset" on NASA's solar eclipse maps? [Answer]
What are those U1, P1, P2, ... on NASA's solar and lunar eclipse maps? [Answer]
How is eclipse duration related to the gamma value of the eclipse? [Answer]
What is the principle behind Einstein's experiment during a total solar eclipse to confirm his general theory of relativity? [Answer]
Is it possible to project the number and approximate form of the Bailey's beads during a solar eclipse? [Answer]
The theoretical maximum is 1 hour 47 minutes, and the total (lunar) eclipse on 16 July, 2000 nearly approaches this limit. The following criteria should be met to create a long total lunar eclipse:
The lunar eclipse should occur in the (northern hemisphere) summer months so that the Sun-Earth distance is longer and the Earth's umbral shadow can extend more to the other side.
The Moon should be near its apogee (i.e. farthest away from Earth). Although this will make the Moon travel across a smaller cone of shadows , the advantage is that the Moon travels the slowest in apogee, due to Kepler's second law of planetary motion which states that the line joining a revolving object and the focus that the mass is at sweeps out equal areas in equal periods of time.
The Moon should reach deep in the Earth's umbra, the best possible case being the centre of the Moon passing the Earth's shadow axis. [Back to top]
The eclipsed sun during totality is safe to watch with naked eyes without filters. The eclipsed disk is surrounded by faint patches of light known as the solar corona (the solar atmosphere). The sky is dark enough for some bright stars, illuminated planets and comets to appear, but is not as dark as that of a night sky. The lowest light intensity is about 2 lux, about the intensity of a night sky with a full moon. The colour is orange or yellow all around the horizon, as if there were sunrise/sunset everywhere. This is due to the small size of the lunar shadow - the light you see near the horizon comes from places where the Sun is not totally eclipsed at that moment.
The following events can be expected, with their respective times given in the first column. Note that some of the events below can occur simultaneously, and the following assumes that the observer is close to the central line of the eclipse:
When 80% of the Sun is eclipsed; about 15-20 minutes before totality
- Sky starts to darken and appears increasingly different from the brightness of an overcast sky
5 minutes before totality
- Moon's (very dark) shadow comes from the west
- Animals start to behave strangely. Some may interpret this as sunset and act accordingly.
1 minute before totality
- Shadow bands appears on bright surfaces
- Temperature drops significantly
10-20 seconds before totality
- Diamond ring appears
1-10 seconds before totality
- Bailey's beads form and vanish at different locations due to the Moon's rough surface
- Twilight sky detected all over the horizon
- Corona appears
- Bright stars, planets and comets may appear
- Temperature continues to drop
- Prominences may be visible in the first and last stages of totality. They will be visible for longer at places far from the central line.
1 minute before totality ends
- Sky begins to brighten in the west and darken in the east
about 10-20 seconds before totality ends
- Prominences appear again, this time on the western edge of the Sun
- Corona starts to fade
- A beam of light returns to the observing site. The sky suddenly brightens.
1-10 seconds after totality
- An increasing number of Bailey's beads form, eventually joining themselves into a ring
- The corona fades more significantly
- Moon's shadow rushes to the east
about 30-60 seconds after totality
- Shadow bands may be seen on bright surfaces
- Animals resume normal activities
- Temperature starts to rise significantly
Thereafter, the partial eclipse occurs in reverse order until the Sun is no longer covered by any part of the Moon. [Back to top]
The Saros is a cycle in which after a cycle has elapsed, eclipses with similar characteristics will recur. Each Saros lasts for 6585.3 days (i.e. 18 years 11 days and 8 hours) and is roughly the common multiple of the following:
The synodic month (related to the lunar phase): 29.53 days - One Saros is equal to 223 lunations
The draconic month (from node to node): 27.21 days - One Saros is equal to 242 passages of a node
The anomalistic month (from perigee to perigee): 27.55 days - One Saros is equal to 239 passages of the perigee
Hence one would expect that the three heavenly bodies will align in a similar configuration after a Saros period has passed, and will produce a similar eclipse. This cycle is believed to be first discovered by the Babylonian Chaldeans several hundred years before common era begins. Notice that ancient astronomers did not necessarily know that an eclipse they predicted had in fact occurred without their notice, simply because it occurred in a different place. Hence it must have been a difficult task for them to discover these theories and eventually a cyclic pattern before the emergence of modern celestial mechanics calculation.
This cycle is extremely convenient for people to predict eclipse occurrences because of its extreme accuracy and simplicity - after an eclipse has occurred, add 6585.3 days and it is very likely for another eclipse to occur on the date computed. However, as noted in the previous paragraph, not every eclipse will be visible in a particular place, but still this is a very elegant way to predict eclipse occurrences.
Although eclipses in the same Saros family share similar characteristics and adjacent Saros members (e.g. Eclipses A and B where B occurs 6585.3 days after A) often resemble one another, there are still some differences that make accurate prediction in terms of eclipse visibility impossible without further calculations:
Adjacent Saros members are separated by 6585.3 days. This means that an extra 8 hours will need to pass and thus the Earth will have rotated for another 120 degrees. This causes the next eclipse to be visible about 120 degrees longitude to the west, and places that enjoyed totality/annularity in the previous visit will almost always lie outside the track in the next eclipse in the series. For this reason, eclipses that are separated 3 Saroses apart (i.e. 19756 days or 54 years and 34 days) will be visible in approximately the same longitude. However this still does not guarantee eclipse visibility due to the factors to be mentioned below.
The fact that the Saros cycle lasts 18 years and 11 days means that the next eclipse will occur later in the year. This will make the relative sizes of the Sun and the Moon different and that the track will be shifted north or south depending on which node the Moon is passing at the time of eclipse. For example, if an eclipse occurs on 1st September, the eclipse after one Saros will be on 12th September, and the Sun will be a little bit closer to us (perihelion occurs in January and aphelion in July). There is a chance that the Moon cannot cover this "larger" Sun and the Saros may cease creating total eclipses.
Moreover, this common multiple relationship is only approximate - 223 lunations and 242 (ascending) node passages do not consume the same number of days. Thus after one Saros cycle the Moon is not at the same position with respect to the nodes. The result is that each Saros will progress from (non-central) partial eclipses to central eclipses, then to partial eclipses again until the Moon is off a node by so much that eclipses cease to occur (i.e. the Saros ends). This whole process takes about 1300 years, and a Saros series has about 72 members. [Back to top]
Four of the 71 solar eclipses in Saros 136. Note that Hong Kong got pretty decent partial eclipses in both 1955 and 2009 - a separation of 3 Saros cycles (The actual date separation is 54 years 32 days because of extra leap years during that period).
Saros are numbered according to the date on which the "middle" eclipse of that series occurs (i.e. the eclipse that is closest to the ascending/descending node, or equivalently the eclipse in which the lunar shadow axis is closest to the Earth's centre), not based on the date that the Saros starts. The following table shows a few consecutive Saros series (I used solar eclipses below):
Date of middle eclipse
Date of the commencement of that Saros
For solar eclipses, eclipses from odd-numbered Saroses occur at the ascending node, and they migrate from the north pole to the south pole in their "lives"; those from even-numbered Saroses occur at the descending node and migrate from the south to the north.
For lunar eclipses the numbering is opposite to that for solar eclipses. Odd-numbered Saroses generate eclipses at the descending node and progress northward with respect to Earth's axis while even-numbered Saroses generate eclipses at the ascending node and progress southward with respect to Earth's axis. [Back to top]
For solar eclipses, Saros 117 to Saros 156 (inclusive) are active, with Saros 117 ending on 2054/08/03 and Saros 157 coming into play on 2058/06/21. For lunar eclipses, Saros 110 to Saros 150 (inclusive) are active, with Saros 110 ending on 2027/07/18 and Saros 151 coming into play on 2096/06/06. [Back to top]
During some total lunar eclipses it may be possible to observe a dark red colour instead of the expected disappearance of the Moon. This is because although the Moon receives no direct light rays from the Sun, the Earth's atmosphere can bend light from the Sun to reach the Moon. Colours except red are mostly filtered by the Earth's atmosphere, and consequently most light that is refracted will appear red or orange and is dimmer than the normal light. Hence it will be possible to observe some colour on the Moon although it is totally eclipsed.
Various conditions on the Earth, especially at places that are responsible for the bending of light, will create different lunar appearance during total lunar eclipses. More dust particles, perhaps as a result of volcanic eruptions, will usually block more light and consequently allow less to reach the Moon, thus the Moon will appear darker (in dark red or brown) or even invisible.
The Danjon scale is proposed by the French astronomer Andre-Louis Danjon to describe the Moon's appearance during a total lunar eclipse, the classification of which is shown below:
Danjon Scale Value (L)
Description of Moon's appearance
Very dark eclipse. Moon almost invisible, especially at mid-totality.
Dark Eclipse, gray or brownish in coloration. Details distinguishable only with difficulty.
Deep red or rust-colored eclipse. Very dark central shadow, while outer edge of umbra is relatively bright.
Brick-red eclipse. Umbral shadow usually has a bright or yellow rim.
Very bright copper-red or orange eclipse. Umbral shadow has a bluish, very bright rim.
The maximum width of the track of annularity/totality is about 200 to 300 kilometres near the tropics, but can be as wide as 500 kilometres or above at higher latitudes due to the curvature of the Earth. In most cases the track runs from west to east because the Moon itself travels from west to east which is in the same direction as the Earth's rotation. However, exceptions exist for eclipses that take place near the poles. In such cases it may be possible for the track to run from east to west when the shadow spot is beyond the pole (to the other hemisphere which should be in the dark - except for places very close to the poles where they can receive light for 24 hours a day).
Total eclipses with very large magnitudes and annular eclipses with small magnitudes usually have wider widths of totality/annularity because such phenomenon can be observed in more places. This explains why hybrid eclipses usually have very narrow tracks of totality/annularity. [Back to top]
In the annular eclipse on 2008/02/07 near the South Pole, the shadow actually travels from east to west for some time before it eventually goes north and returns to the normal west-to-east movement.
The shadow moves at about 3400 km/h (0.94 km/s) with respect to the Earth's centre. However, due to orbital motion of the Earth, the shadow moves the slowest near the equator with respect to the observer because the Earth's rotation is fastest there. The speed is approximately 1700 km/h (0.47 km/s). However it can be much faster near the poles and when the eclipse starts or ends where it runs down the curvature of the Earth. At this stage the speed can be as high as 30000 km/h (9 km/s). [Back to top]
No. The path of totality or annularity can pass through any place on Earth. For instance, a total and annular eclipse is visible at North Pole on 2015/03/20 and 2021/06/10 respectively; while the South Pole will enjoy totality and annularity on 2094/01/16 and 2097/11/04 respectively. (All dates are in accordance with UTC times) [Back to top]
The four solar eclipses mentioned above.
They define the boundaries where eclipses are visible/not visible at certain times. Explanations are given as follows:
Eclipse finishes at sunrise : At places on that line, the solar eclipse has already finished while the Sun is below horizon; the Moon leaves the solar disk at the point of sunrise, and that places further away from that line will not see any solar eclipses.
Maximum eclipse at sunrise : As the name suggests, for places on that line the eclipse reaches maximum at sunrise, and all subsequent phases can be seen there. Thus those places will enjoy half of the eclipse process.
Eclipse starts at sunrise : For places on this line, the whole eclipse process can be seen, with the Moon entering the solar disk just at sunrise.
i.e. Places between  and  will see less than half of the eclipse, and naturally greatest eclipse for such places will be at the time of sunrise; places between  and  will see more than half of the eclipse including maximum eclipse. Thus for those places the time of greatest eclipse will be after sunrise.
Eclipse finishes at sunset : Observers at that line will just be able to see the full eclipse process before the Sun dips below the horizon.
Maximum eclipse at sunset : Observers there will see half of the eclipse; the sun sets when maximum eclipse occurs.
Eclipse starts at sunset : Observers there will not see any eclipse phases, as the eclipse only starts when the Sun is at the horizon.
i.e. Places between  and  will see less than half of the eclipse, and naturally greatest eclipse for such places will be at the time of sunset; places between  and  will see more than half of the eclipse including maximum eclipse. Thus for those places the time of greatest eclipse will be before sunset. [Back to top]
For solar eclipse:
P1 denotes the earliest time when the lunar penumbra touches Earth. This marks the start of partial eclipse on Earth.
P2 denotes the earliest time when the whole lunar penumbra is cast on Earth. Not all eclipses have this phase because it is perfectly possible that some of the penumbra is cast into space during the whole eclipse process.
P3 denotes the latest time when the whole lunar penumbra is cast on Earth. Again not all eclipses have this phase because it is perfectly possible that some of the penumbra is cast into space during the whole eclipse process.
P4 denotes the latest time when the lunar penumbra touches Earth. This marks the end of partial eclipse on Earth.
These labels are similarly defined for umbral/antumbral contacts. Note that purely partial eclipses will not have any of the U labels as the lunar umbra/antumbra never touches Earth:
U1 denotes the earliest time when the lunar umbra/antumbra touches Earth. This marks the start of total/annular eclipse on Earth.
U2 denotes the earliest time when the whole lunar umbra/antumbra is cast on Earth. Not all total/annular eclipses have this phase because it is possible that some of the umbra/antumbra is cast into space during the whole eclipse process (this is true for non-central total/annular eclipses which are quite rare).
U3 denotes the latest time when the whole lunar umbra/antumbra is cast on Earth. Again not all total/annular eclipses have this phase because it is possible that some of the umbra/antumbra is cast into space during the whole eclipse process.
U4 denotes the latest time when the lunar umbra/antumbra touches Earth. This marks the end of total/annular eclipse on Earth.
For lunar eclipse:
P1 denotes the time when the Moon first touches Earth's penumbra (external tangency). Penumbral eclipse starts.
U1 denotes the time when the Moon first touches Earth's umbra (external tangency). Partial eclipse starts.
U2 denotes the time when the Moon is first fully inscribed in Earth's umbra (internal tangency). Total eclipse starts.
U3 denotes the latest time when the Moon is fully inscribed in Earth's umbra (internal tangency). Total eclipse ends.
U4 denotes the time when the Moon is fully out of the umbra (external tangency). Partial eclipse ends.
P4 denotes the time when the Moon is fully out of the penumbra (external tangency). Penumbral eclipse ends. [Back to top]
In a solar (lunar) eclipse, the gamma value is the minimum distance between the Moon's (Earth's) shadow axis and the centre of the Earth (Moon), the unit being Earth's radius. This means that eclipses with smaller (absolute of) gamma values are generally closer to the centre of the object eclipsed. In solar eclipse, this means that (in general) the umbral/antumbral shadow will stay longer on Earth, creating a longer path of totality/annularity. This effect is also brought by the fact that solar eclipses with a lower (absolute of) gamma values occur in lower latitudes, and for those places the speed of the Earth's rotation is the fastest, thus bringing a longer eclipse. For lunar eclipses, eclipses having smaller (absolute of) gamma values occur when the Moon is deep within Earth's shadow, hence the duration of totality is longer. [Back to top]
One of the consequences of the general theory of relativity is that masses can bend space and time, and light can also be bent by any mass. Since the Sun is very massive (2 x 1030 kg), it will also bend light reaching us from other stars if the theory is correct. Unfortunately the Sun is so bright that normally stars close to the Sun are invisible, with only one exception - when the sun is totally eclipsed. The ingenious Einstein thus proposed to measure the apparent angular separation between two stars that are close to the Sun during a total solar eclipse, and compared with the distance between them half a year later, when they would be visible in the night sky. According to his (revised) calculations, the deflection should be 1.75 arc seconds.
Scientists went to Principe, just off the west coast of Africa, to take their measurements during the total solar eclipse on May 29, 1919. Comparing with night-sky observations of the same two stars, scientists found a difference very close to his prediction, and this eventually contributed a piece of evidence to Einstein's great theory. [Back to top]
Bailey's beads are always different in each eclipse, and from place to place. However, by situating oneself near the southern or northern limits of the path of totality, one will be able to observe more Bailey's beads because the Sun will just be covered by the Moon. It is also possible to estimate in advance the properties of Bailey's beads using the lunar limb profile, one example of which is shown below:
Using this diagram, one will be able to predict the topographic features of the Moon (which are exaggerated so that they can be seen easily in the map) during an eclipse, and predict where and roughly how many Bailey's beads can be seen. [Back to top]