The latest thing going around the internet today is that the Sun rose on in a small burg called Ilulissat in Greenland on January 11th. Ilulissat is north of the arctic circle so it experiences 24 hours of darkness for several weeks. Nothing unusual there.
The unusual part is that the Sun usually rises again on January 13th. Therefore, the Sun came back two days early and the question is why.
The first thing to do is to verify the claims is true. I checked out the NOAA Sunrise/Sunset calculator (after finding what I think is the lat/long of Ilulissat. There are lots of places with similar names in Greenland!) and the US Naval Observatory. Sure enough, January 12th, no sunrise. January 13th, there is a sunrise and a very short day. Okay, NOAA and USNO agree with the prediction and they are pretty reputable in such matters.
The next thing is to verify the observation. This part is much trickier. The story has gone viral on the web. The best I can tell, it started with a story in the Daily Mail. Read it and you can see the details of who made the observations and where are pretty sketchy...okay, non-existant. No mention of the time of the observation, how long the Sun stayed up, location of observation, azimuth of the Sun (I don't consider the stock photo they used in the story to have any bearing on anything) or who made the observation. A photo of the sunrise with an identifiable landmark would have been REALLY useful!
Now on to the possible why (assuming it really happened).
First, let's get rid of all the crackpot theories. Earth's rotational axis has not changed. Astronomers use telescopes every night and measure the positions of stars down to tenths of an arc second (or finer) resolution every night. If something had happened like that, everyone around here would be totally whigged out about now, yet calm prevails at observatories around the world. This is not related to the bird deaths, 2012 non-sense, magnetic pole shifting, etc.
The idea that is getting the most play and traction is that global warming is to blame. Greenland has been well above average this year by about 3 degrees Celsius. The running idea is that the glaciers are melting and the horizon is lower. A lower horizon lets the Sun peak above it a little earlier. If this is the case, surveying crews should be able to make some measurements and confirm or deny this rather quickly.
I decided to play around a little bit and see what I could find out. Checking out the NOAA Solar Position Calculator , I found the center of the Sun was only 0.68 degrees below the horizon at local noon. Since the Sun has a radius of about .27 degrees, the center of the Sun was only .41 degrees below the horizon. Converting to radians gives an angle of about .007 radians. So now I have a pretty good idea of the the angle the tip of the glaciers must have descended to let the sun rise two days early.
But how many meters of melting would have to take place? The math is easy, but knowing the numbers to use is difficult. The answer depends very much on how far away these glaciers are located. Looking at Google Earth and Google maps is of limited utility since you I would need to know exactly where they were looking!
However, I can still make some general calculations here. Let's say the horizon is 10,000meters away (about six miles). The height the glaciers must have lost is simply the angle times the distance to the mountains (using the small angle approximation here and the angle is small enough that works pretty well) or 10,000*.007 = 70 meters or about 220 feet! Of course, if the glacier is much closer, the amount of ice lost could be much smaller. If you were 1000 meters away, it would only need to lose 7 meters (about 22 feet) off the top. However, without information on exactly where the observation was made, there's not much you can say about it (although that hasn't stopped any of the whackos out there!)
Now I should also point out that it could also be due to the ground rising. As glaciers melt, the ground will rise as there is less weight on it. I doubt this could rise enough in one year to create the effect by itself, but if the glaciers fell a little and the ground rose you might be in business.
However there are other explanations. Accuweather proposed an unusual concentration of methane in the atmosphere may have changed the index of refraction of the atmosphere at let the Sun be visible a couple of days early. I find this implausible as the amount of methane needed to do this would be huge! Volcanic eruptions can release enough pollutants into the atmosphere to give colorful sunsets, but even they do not change the atmosphere enough to affect the time of sunsets!
Much more likely is an atmospheric phenomena. A certain rare but understood shape of ice crystal (called a pyramidilly-terminated hexagonal prism for those following along at home) can produce a halo with a radius of 4 degrees. This halo would extend a little above the horizon giving the illusion of a sunrise. I think it is fairly safe to say that Greenland is cold enough for atmospheric ice crystals to form at this time of year. These are pretty rare but then again this doesn't happen every year! Again, a good picture of this would be amazingly useful.
Although I need more info, I would put my money on the small halo. The amount of ice that would have to fall off the top of a glacier is pretty large and a fall of that size probably would have been noticed so let's see some evidence of it (I know this just happened and in a pretty remote place, so I am willing to give them a little time to get the info here).
As regular readers know, I love photographing sunsets...maybe someone will fund me to go to Greenland next January and take some pics to get to the bottom of this!
Reprinted with permission from the Half-Astrophysicist Blog.