MIT Infinite Corridor Astronomy - MIThenge

Originally prepared by Ken Olum. Maintained by Keith Winstein.

CAES has decided to maintain their own MIThenge web page. The CAES MIThenge page includes predictions of MIThenge for the next 100 years.

As viewed from a stationary point on the earth, the path of the sun through the sky is roughly a circle which moves north and south as the seasons go by. In mid-November and in late January every year, the circular path crosses the axis of MIT's Infinite Corridor, which runs a distance of 825 feet (251 meters) from the main entrance on Massachusetts Avenue through Buildings 7, 3, 10, 4 and 8. When this happens, the setting sun can be seen from the far end of the corridor. By analogy with Stonehenge, this phenomenon is sometimes called "MIThenge". (The same cannot be seen at sunrise because the other end of the infinite corridor is blocked by Building 18.)

Pictures

Here are three photographs of this phenomenon, the first from approximately a third of the way down the corridor on the third floor, and the other two from the entire distance on the second floor. Click for full-size versions.

January 2001. Photograph by Matt Yourst November 1999. Photograph by Joseph Kaye
If you have more photographs of the sun through the length of the infinite corridor, please send them to me if they're in electronic form, or contact me if you want me to come by to scan them. If your photographs are precisely dated and timed, even better: this helps our understanding of the corridor's azimuth and the light-bending effects of the atmosphere needed to correctly predict when MIThenge will occur. There appears to be a systematic error in the calculations on the order of 5-8 minutes that I'm trying to track down.
February 2, 1998. Photograph by Paul Schechter, courtesy Rainer Weiss

2001-2002 Predictions

The predictions shown below give the times at which the center of the sun will cross the azimuth of the infinite corridor axis, and the observed altitude of the center of the sun above a level reference plane at that time, in degrees:minutes:seconds of arc. The sun is approximately 30 minutes of arc (half a degree) in diameter, and so extends about 15' in each direction. Because there is a hill in Boston that blocks a level view along the corridor, only parts of the sun that are somewhat above the level plane are visible. There is concern that the infinite corridor may actually be tilted on the order of 1 degree with the west side higher, which if true is not currently taken into account. Caveat spector. 


Date            Time            Altitude
11/10/2001      16:17:49        1:06:27
11/11/2001      16:19:05        0:45:46
11/12/2001      16:20:20        0:26:00
11/13/2001      16:21:36        0:06:54
11/14/2001      16:22:51       -0:11:12


Date            Time            Altitude
1/27/2002       16:50:45       -0:04:46
1/28/2002       16:49:50        0:13:46
1/29/2002       16:48:53        0:33:12
1/30/2002       16:47:53        0:53:38
1/31/2002       16:46:53        1:14:35

Warning

The fact that the sun is at the horizon does not mean that it is safe to look at for more than a moment. Do not injure your eyes by staring at the sun. Use appropriate filters or glance only for a moment as you would at the sun up in the sky.

What you'll see

As the sun becomes better and better aligned with the corridor the amount of floor that is illuminated goes up and up. Since the floor is very reflective this means that the reflected sun can be seen far down the corridor as the event approaches. The orange light reflected onto the ceiling is often striking. This phenomenon is visible for a week or two around the actual sunset days.

At some point, some portion of the sun's disc will be visible from the far end of the corridor. Which part of the disc will be seen first depends on whether the sun is crossing above or below the corridor. If the altitude in the table above is high, the sun will cross the upper right edge of the corridor as seen from the lower left. If the altitude is low, the sun will cross from the center left of the corridor and set below the hill.

The disc of the sun is close to the area of sky visible from the end of the corridor. Therefore if your timing is right and you get a good vantage point you will see almost nothing of the sky except for the interior of the sun's disc.

Observing this event depends strongly on atmospheric conditions. I think the best is a very red sunset. This decreases the total light intensity and makes it possible to see the corridor and the sun at the same time. If the sky is brilliantly transparent right down to the horizon then the setting sun will be too bright to look at. When you use a filter to save your eyes (see below) you won't be able to see much of the corridor.

The period of time during which the sun can be seen from the corridor end is quite short -- no more than two minutes, and only as long as that under ideal conditions.

Some suggestions:

View from the Building 8 end.
Don't go up to the Mass. Ave. end of the corridor. You won't see anything special there, and you will prevent others from seeing.
Arrive early and don't block the corridor.
Especially given the apparent systematic tendency for our predictions to be 5-8 minutes late, it's best to arrive 15 minutes early. This helps make sure you're not trying to scramble for a spot and blocking somebody else's view. Above all, do not walk in the corridor in front of all the people trying to catch a glimpse of the sun.
Use the third floor.
The 3rd floor has the best view because it has an uninterrupted sweep of corridor, and because it is the highest usable floor. You can't see anything from the fourth floor because it doesn't go through.
Let others see.
Not very many people can get a good view of this phenomenon at once. Once you have seen, get out of the way and let others look.
Stand in the stairwell.
The place that has the largest unobstructed angle of view of the sky is the stairwell just below the third floor. Get your eye right by the edge of corridor floor. Great view? Now give someone else a chance.
Don't hurt your eyes
See the warning above.

Moon Illusion and Other Unrelated Effects

Ken Olum has found that the infinite corridor produces the "moon illusion", in which the setting sun looks gigantic in the sky, even though it is no larger at the horizon than it is at the zenith. On an unrelated point, note that the size of the moon does indeed vary by approximately 12 percent between perigee and apogee. For more information, see John Walker's "Inconstant Moon" page, which includes the following (again unrelated to MIT or the Infinite Corridor) to-scale diagram (650 km/pixel) of the Earth-Moon system:

Data, Computations, and Sources of Error

These predictions depend on knowing the azimuth of the infinite to an accuracy of several arc-minutes. The best figure we have found is the accurately surveyed azimuth of MIT's property line along Memorial Drive, which is given as 245.47391075 degrees true. (Note that at this precision, an error of 1 in the last decimal place corresponds to a crosswise error of 438 angstroms over the length of the infinite corridor, or to shooting a bullet from the earth at the center of the sun and missing the center by 29 yards.) According to Michael K. Owu '86 of the (apparently now-defunct) MIT Planning Office, the Infinite Corridor should have been constructed parallel to this line. Manual measurements on the adjacent 0.5 meter resolution orthophoto (courtesy MIT Course 11's ortho.mit.edu and MassGIS) come up with essentially this same number within their precision, which alone would be good enough to expect around a minute of error in predicted times. Other possible causes of error include computer program bugs, problems with the atmospheric model, carelessness, and a suspected non-levelness of the infinite corridor.

The predictions are obtained today with xephem (available on Athena), the given azimuth, a GPS-surveyed lat/lon for the East side of the infinite corridor, and a guess at the barometric pressure. None of these factors except the azimuth matters much to the final result, and the figures thus obtained tend to be identical (to the second) to what Ken Olum has previously predicted using unknown means. My current hunch is that our understanding of the infinite corridor's slope may be off. On the day Professor Schechter shot his above photograph (2/2/98), the sun's altitude was 1'58'' (almost 2 degrees) when it was at the right azimuth (4:44:48, with the photograph marked as "4:45"), with very little dependence (+/- five arcminutes) on temperature and barometric pressure. By our predictions (with the infinite corridor approximately 9 feet by 13 feet by 825 feet), this should not even have been visible.

Here is a graph showing the path of the center of the sun and its lower edge the day the picture was shot, using the same methodology used to predict MIThenge events. As you can see, the Sun disc comes nowhere near a level infinite corridor. The azimuth appears to be correct and confirmed by the "4:45" notation from the photograph, but the elevation seems to be way off. (Note: the plot says, "Time extent: 4:40 to ..." but the bottom of the Sun disc does not actually start to show up on the plot until 4:42:57.) If indeed the infinite corridor is tilted up as much as would be necessary to see this predicted sun, that would mean the floor in building 7 is higher (in ellipsoid height) than the ceiling in building 8. This seems hard to believe.

You can help improve the predictions by taking sightings during the event. Especially valuable are photographs from the length of the corridor with a precise timestamp, particularly if they show only half of the doorway on the far side of the corridor filled with the sun. If the entire doorway is filled with the sun, that's okay too, but then we don't know what part of the sun you ended up photographing (middle, side, etc.) I am not an expert in solar photography; you should make sure not to damage your camera.

Up to MIT Planning Office

Up to MIT

keithw@mit.edu