A commenter on reddit points out that the Winter Solstice itself may be pretty arbitrary. Too true!
So, let’s consider another possibility, and calculate from perihelion to perihelion. According to the US Naval Observatory, the last perihelion was at 3pm UTC on January 3, 2010, and the next will be at midnight on January 3, 2011. Turning again to Frink, that would give us a date of April 29, 2010 at around 3pm UTC.
#2010-01-03 15:00 UTC#+((#2011-01-03 00:00 UTC#-#2010-01-03 15:00 UTC#)/pi)->UTC
Any further refinements are appreciated!
To an amateur astronomer, this sounds like the most reasonable choice. Many of the great observatories are based in the Southern Hemisphere, and so to discriminate against those fine contributors to our body of knowledge is thoughtless.
There are many ways to calculate such an event, but because of the basis in science of this particular one, lets be as objective as possible.
Dave Lawson
Actually, the appropriate point is the first point of aries, (i.e. the vernal equinox) as that is the 0,0 point in the celestial coordinate system.