APOD: Full Moon in Earth's Shadow (2015 Apr 08)

[APOD Robot]

Full Moon in Earth's Shadow

Explanation: Last week the Full Moon was completely immersed in Earth's dark umbral shadow, just briefly though. The total phase of the April 4, 2015 lunar eclipse lasted less than 5 minutes, the shortest total lunar eclipse of the century. In fact, sliding just within the Earth's umbral shadow's northern edge, the lunar north stayed relatively bright, while a beautiful range of blue and red hues emerged across the rest of the Moon's Earth-facing hemisphere. The reddened light within the shadow that reaches the lunar surface is filtered through the lower atmosphere. Seen from a lunar perspective it comes from all the sunsets and sunrises around the edges of the silhouetted Earth. Close to the shadow's edge, the bluer light is still filtered through Earth's atmosphere, but originates as rays of sunlight pass through layers high in the upper stratosphere. That light is colored by ozone that absorbs red light and transmits bluer hues. In this sharp telescopic view of totality from Auckland, New Zealand, planet Earth, the Moon's north pole has been rotated to the top of the frame.

<< Previous APOD

This Day in APOD

Next APOD >>

[/b]

[Guest]

There's a typo in the full resolution jpg link on the description page.

http://apod.nasa.gov/apod/image/1504/TLE2015_TLE2015_3239x2597olsen.jpg
should be
http://apod.nasa.gov/apod/image/1504/TLE2015_3239x2597olsen.jpg

[Nitpicker]

In this sharp telescopic view of totality from Auckland, New Zealand, planet Earth, the Moon's north pole has been rotated to the top of the frame.

It is a lovely image, but the Moon's north pole is about 20° clockwise from up. (And it is of course a mirror image.)

[alter-ego]

... The total phase of the April 4, 2015 lunar eclipse lasted less than 5 minutes, the shortest total lunar eclipse of the century.

Barely total or barely partial, we may never know.

[Boomer12k]

Awesome pic!!

:---[===] *

[Unixwolf]

Many years ago I lived in the little village of Cornwallis, about 20 km west of Auckland. One evening it seemed to me that there was something wrong with the moon. I suddenly realized what it was. The first thing I did when I got to work the next morning was open up the source to "Phoon" <http://acme.com/software/phoon/>, change the sign of a constant, and recompile it. Now the picture on my terminal looked just like the moon as seen from the my backyard.

[Steve Dutch]

The moon isn't just rotated, but inverted right-left. Did someone merely flip the moon instead of rotate it?

When I first saw the moon from the Southern Hemisphere I was amazed to see it "right side up:" south up, the way I was used to seeing it in a telescope.

The hype over this "shortest" eclipse made it sound like you'd miss it if you blinked. In reality, someone at the moon's north pole saw the Sun hidden for only five minutes, but someone near, say, Copernicus, saw totality lasting a couple of hours. Terrestrial observers saw a normal eclipse. In my location, we had an eclipse of the whole sky (clouds).

[Nitpicker]

The moon isn't just rotated, but inverted right-left. Did someone merely flip the moon instead of rotate it?

When I first saw the moon from the Southern Hemisphere I was amazed to see it "right side up:" south up, the way I was used to seeing it in a telescope.

The hype over this "shortest" eclipse made it sound like you'd miss it if you blinked. In reality, someone at the moon's north pole saw the Sun hidden for only five minutes, but someone near, say, Copernicus, saw totality lasting a couple of hours. Terrestrial observers saw a normal eclipse. In my location, we had an eclipse of the whole sky (clouds).

If there are an odd number of mirrors in the optical train, a standard camera will always record a mirror image, regardless of the rotation. Some people just decide not to un-mirror the final image. And to refine my earlier comment, I'd say the Earth's north pole has been rotated to point (very close to) up in this image, not the Moon's. In other words, celestial north is up and celestial east is to the right (whereas it is conventionally to the left).

I think an observer on the Moon's north pole, might just have seen a sliver of Sun at maximum eclipse. The boundary between total and partial is a little fuzzy when the Earth's atmosphere is the deciding factor.

[eekee]

404 on clicking the image.

other than that, it's a lovely image

[eekee]

never mind; shift-clicking the reload button got a fixed link.

[Joe Stieber]

... The total phase of the April 4, 2015 lunar eclipse lasted less than 5 minutes, the shortest total lunar eclipse of the century.

Barely total or barely partial, we may never know.

Sky & Telescope has an online article about the questionable totality...

http://www.skyandtelescope.com/astronom ... -04062015/

[Chris Peterson]

Barely total or barely partial, we may never know.

I'm not sure about that. It depends on definitions. Normally, we would ignore the effects of the atmosphere, and simply consider the geometry. That can be done rigorously.

I suspect this was actually a total eclipse from a geometrical standpoint, but appeared partial because of atmospheric effects.

[MarkBour]

Barely total or barely partial, we may never know.

I'm not sure about that. It depends on definitions. Normally, we would ignore the effects of the atmosphere, and simply consider the geometry. That can be done rigorously.

I suspect this was actually a total eclipse from a geometrical standpoint, but appeared partial because of atmospheric effects.

Solar prominences are not entirely rigid, however.

[Chris Peterson]

Barely total or barely partial, we may never know.

I'm not sure about that. It depends on definitions. Normally, we would ignore the effects of the atmosphere, and simply consider the geometry. That can be done rigorously.

I suspect this was actually a total eclipse from a geometrical standpoint, but appeared partial because of atmospheric effects.

Solar prominences are not entirely rigid, however.

Agreed. But they're ignored in a geometrical model, as well. And I doubt they would be bright enough to impact the appearance of the Moon as seen from Earth.

[William Foster]

So that (umbral shadow) would be a shadowy shadow? From the latin "umbra" = shadow. This is the
origin of "umbrella"
Thanks for the site and its riches!

[alter-ego]

Barely total or barely partial, we may never know.

I'm not sure about that. It depends on definitions. Normally, we would ignore the effects of the atmosphere, and simply consider the geometry. That can be done rigorously.

I suspect this was actually a total eclipse from a geometrical standpoint, but appeared partial because of atmospheric effects.

Well, I'm certainly in agreement with you about not being sure.
Interestingly, the visual umbra is counterintuitive. In fact the geometric umbra is smaller by ~2%, not larger as I believe you inferred. The official eclipse predictions we see include an average correction for atmosphere perturbation of the geometrical umbra. Prediction variations in eclipse magnitude (depth) depend on the method use to estimate umbral size (not standardized). Fred Espenak does the eclipse calculations for the NASA Eclipse Website. He describes the histories and descriptions of the two methods for determining the size of the umbra are here. The accepted definition of the visual umbral edge is the point of inflection in the shadow density (steepest intensity gradient). Also, I'll add that crater timings have been and are critical for assessing umbral size and variations.

This eclipse is a rare one; whether it is barely total or deep partial depends entirely on the atmosphere and possibly 2nd order corrections to Earth's actual shape, not the geometrical umbra. Just to be clear, whether this eclipse is total or partial is fun to think about, but Fred's comment below reflects my question of whether we'll really know the true depth of this eclipse.

So the small magnitude differences discussed here are only of academic interest. Still, it is important to note which shadow enlargement convention is used because it is critical in comparing predictions from different sources. In both the Thousand Year Canon of Lunar Eclipses 1501 to 2500 and the Five Millennium Catalog of Lunar Eclipses: -1999 to +3000 (NASA TP-2009-214173), Earth's penumbral and umbral shadow sizes have been calculated by using Danjon's enlargement method. Other sources using Danjon's method include Meeus and Mucke (1979), Espenak (2006) and Connaissance des Temps. Several sources using Chauvenet's method are Espenak (1989), Liu and Fiala (1992), and Astronomical Almanac.

[Banjo-Jim]

Greetings from Argentina. The place where everything is upside-down!

So why is APOD of the moon 8 April 2015 bass ackwards? ;>)
Banjo-Jim

[Nitpicker]

Barely total or barely partial, we may never know.

I'm not sure about that. It depends on definitions. Normally, we would ignore the effects of the atmosphere, and simply consider the geometry. That can be done rigorously.

I suspect this was actually a total eclipse from a geometrical standpoint, but appeared partial because of atmospheric effects.

Well, I'm certainly in agreement with you about not being sure.
Interestingly, the visual umbra is counterintuitive. In fact the geometric umbra is smaller by ~2%, not larger as I believe you inferred. The official eclipse predictions we see include an average correction for atmosphere perturbation of the geometrical umbra. Prediction variations in eclipse magnitude (depth) depend on the method use to estimate umbral size (not standardized). Fred Espenak does the eclipse calculations for the NASA Eclipse Website. He describes the histories and descriptions of the two methods for determining the size of the umbra are here. The accepted definition of the visual umbral edge is the point of inflection in the shadow density (steepest intensity gradient). Also, I'll add that crater timings have been and are critical for assessing umbral size and variations.

This eclipse is a rare one; whether it is barely total or deep partial depends entirely on the atmosphere and possibly 2nd order corrections to Earth's actual shape, not the geometrical umbra. Just to be clear, whether this eclipse is total or partial is fun to think about, but Fred's comment below reflects my question of whether we'll really know the true depth of this eclipse.

So the small magnitude differences discussed here are only of academic interest. Still, it is important to note which shadow enlargement convention is used because it is critical in comparing predictions from different sources. In both the Thousand Year Canon of Lunar Eclipses 1501 to 2500 and the Five Millennium Catalog of Lunar Eclipses: -1999 to +3000 (NASA TP-2009-214173), Earth's penumbral and umbral shadow sizes have been calculated by using Danjon's enlargement method. Other sources using Danjon's method include Meeus and Mucke (1979), Espenak (2006) and Connaissance des Temps. Several sources using Chauvenet's method are Espenak (1989), Liu and Fiala (1992), and Astronomical Almanac.

I agree completely, alter-ego. From a purely geometric standpoint, ignoring the atmosphere on Earth, this was a partial eclipse. This is confirmed by my own hand calculations, and perhaps more convincingly by Stellarium, which uses the same VSOP and ELP computer models as Fred Espenak, for predicting the positions and distances of the Sun and Moon. At the time of greatest eclipse in Stellarium, one can clearly see the top of the Sun from the top of the Moon, and vice versa, with no atmosphere modelled on Earth (and most likely with atmosphere, too).

However, it appears that the all-but-official and conventional boundary between a partial and total lunar eclipse, involves a few extra, somewhat arbitrary considerations.

[Nitpicker]

Banjo-Jim, to answer your question with a question, why didn't you ask your question in this thread, instead?:
http://asterisk.apod.com/viewtopic.php?f=9&t=34631

(A better answer to your question has already been posted there.)

[alter-ego]

I agree completely, alter-ego. From a purely geometric standpoint, ignoring the atmopshere on Earth, this was a partial eclipse. This is confirmed by my own hand calculations, and perhaps more convincingly by Stellarium, which uses the same VSOP and ELP computer models as Fred Espernak, for predicting the positions and distances of the Sun and Moon. At the time of greatest eclipse in Stellarium, one can clearly see the top of the Sun from the top of the Moon, and vice versa, with no atmosphere modelled on Earth (and most likely with atmosphere, too).

However, it appears that the all-but-official and conventional boundary between a partial and total lunar eclipse, involves a few extra, somewhat arbitrary considerations.

The detailed umbra gradient difficult to access and simulate. I'd be cautious of a couple things:
1) I don't take the visual silver sliver in Stellarium as meaningful at this level of detail. I'd bet the atmosphere does not enter into their algorithm at all. Stellarium likely uses a fixed factor larger, and the umbra edge gradient probably is designed to look reasonable. I use Stellarium a lot and find in an incredible tool, but there are some things they do approximate at best and wrong at worst. I'm talking about details that most people don't care about, or the situation doesn't require precise accuracy in detail. 2) It may be the visual sliver you and others saw in real life is a good indicator of a partial eclipse, but you may not be seeing enough to the fuzzy edge to confidently identify the inflection point. I.e. you don't know where the real umbra edge is because you don't see enough of it. I'm only speculating here.

I think a best-effort serious assessment of the eclipse depth would take a couple sets of planned data showing photometric radial light curves before the eclipse and at maximum eclipse. Ratioing them might yield the inflection point location.
See my attempt, and problems, at this from similar 1945 data sets:

http://asterisk.apod.com/viewtopic.php? ... lit=+umbra

Morrison's penumbral gradient measurements and predicted gradient

---

[Nitpicker]

I'm not talking about what the Moon looked like from the Earth in Stellarium (that's art, about which I know naught). But that you could see the north pole of the Moon from the north pole of the Sun in Stellarium, and vice versa. One of the things I love about Stellarium is that the observer can be placed in other parts of the Solar System. (Stellarium only [optionally] models the atmosphere of the Earth if the observer is on Earth.)

What I take away from all this, is that the conventional, enlarged boundary of the Earth's umbral shadow, does not necessarily delineate the boundary where direct sunlight is blocked by the Earth. It is rather based on an arbitrary analysis of the how the light gradient appears on the Moon.

[alter-ego]

I'm not talking about what the Moon looked like from the Earth in Stellarium (that's art, about which I know naught). But that you could see the north pole of the Moon from the north pole of the Sun in Stellarium, and vice versa. One of the things I love about Stellarium is that the observer can be placed in other parts of the Solar System.

What I take away from all this, is that the conventional, enlarged boundary of the Earth's umbral shadow, does not necessarily delineate the boundary where direct sunlight is blocked by the Earth. It is rather based on an arbitrary analysis of the how the light gradient appears on the Moon.

Your basically right, but defining a real, "dull edge" shadow boundary as an inflection point in the intensity gradient is not arbitrary. It is a special point which defines a boundary which I believe might have a mathematical foundation, not just a visual empirical one. Using a simplified enlargement factor of 1.02 is at least empirically supported by crater timing comparisons but is static, and seems arbitrary in that sense. The inflection point definition is a dynamic one and will yield eclipse to eclipse umbral size changes.

The fundamental problem is a simple one: How do you define a clean boundary for a fuzzy edge? I don't know the mathematic details but I wouldn't be surprised the "answer" is the inflection point.

[Nitpicker]

Well, maybe arbitrary was not quite the right word. The boundary has to be defined somewhere, and if you don't go for the geometric option, then I suppose the inflection point is the next logical choice.

[alter-ego]

Well, maybe arbitrary was not quite the right word. The boundary has to be defined somewhere, and if you don't go for the geometric option, then I suppose the inflection point is the next logical choice.

I think so. It seems a fuzzy world leads to fuzzy solutions.

[Nitpicker]

Well, maybe arbitrary was not quite the right word. The boundary has to be defined somewhere, and if you don't go for the geometric option, then I suppose the inflection point is the next logical choice.

I think so. It seems a fuzzy world leads to fuzzy solutions.

The thing I don't particularly like is that the inflection point radius is considerably more unpredictable from one eclipse to the next, compared with the pure geometric radius. So, an average inflection point radius is determined from many eclipses, and is used as the basis for determining whether a future eclipse will be total or partial. But this average radius may end up a poor match against the radius actually measured from the future eclipse, which isn't very helpful when the eclipse is on the cusp of total and partial.