What is an irruption? and don't you mean eruption?
The Birder's Dictionary (Cox 1996) defines irruption as: " the periodic movement of
numbers of birds into unusual ranges for a season."  This should not be confused with
eruption which would be movement out of an area.

Why do Snowy Owls periodically make these large movements way
south of their regular range?
This is a complicated question.  Those of you who remember High School biology
or old episodes of Wild Kingdom probably heard that Snowy Owls move south because
of crashes in the lemming population, an idea proposed by V.E. Shelford in 1945.
The neatness of this simple, cause and effect relationship was so compelling that it still
persists in the literature.  There is absolutely no doubt that lemmings play an essential
role in Snowy Owl ecology on the breeding grounds and local distribution of owls
there changes with lemming population shifts.  The problem is, lemming populations
are not distributed evenly across the landscape at a continental scale.  They form a
variable patchwork or mosaic and significant population changes do not appear to occur
on a scale large enough to explain the spectacular region-wide, sometimes continent-
wide irruptions of Snowy Owls (Parmelee 1992).  It is currently speculated that only
weather phenomena have the capacity to cause continent-wide Snowy Owl movements,
but no one has yet defined what kind of weather phenomena these might be.  And blaming
irruptions on the weather will also probably prove to be too simplistic an explanation.

How often do irruptions occur?
Ah, now here's a question guaranteed to start a fist-fight.  Depending on where you live,
the answer is different.  Northern latitudes get over-winter owls more regularly than
more southerly latitudes.  For example, in British Columbia the frequency of winter owl
occurrences is greater than it is in Oregon.  If we look at Audubon Christmas Bird Count
data (which is the only available continent-wide database, like it or not) we find that
there is at least one count that reports Snowy Owls almost every year, though it's not
necessarily the same location every year.  In Oregon, a Snowy Owl may turn up every
couple years, usually along the North Coast or along the Columbia River east of the
Cascades.  But every 9-12 years there's a big, unequivocal irruption, with 100's of birds
many of them turning up as far south as California.  So the question becomes, how many
owls do you need to call what you're seeing an irruption? and that's where the arguments
start.  How can we sort small random shifts from genuine, statistically significant irruptive
events?  I don't know, I skipped that day in Statistics 101.  Statistical analysis by Kerlinger
and Lein suggests that there is no regular repeatable pattern and that the commonly cited short-
cycle pattern (3-4 years) cannot be supported as statistically meaningful and is therefore, at
best, pseudo-periodic.  Not everyone agrees with their analysis and the use of arguably squishy
Christmas Count data only exacerbates the problem.  We can however, identify the major
events that no one argues about for the West Coast.  These were 1897, 1916, 1947, 1955,
1966, 1973, 1977, 1984, 1996 and now 2005.  These are often followed by what are termed
"echo-flights" in subsequent years.

What's going to happen to these transposed owls?
This is another complicated question.  As part of the general assumption that lemming
crashes were causing movements southward, it was assumed that owls would be in
poor health and starving.  This hypothesis has not really been tested for owls in the
east or west.  Owls in the middle of the continent, where irruptions are more frequent
do not support the "starving" hypothesis.  45% of the specimens examined had moderate
to heavy fat deposits, and traumatic injuries were the major cause of mortality. Causes
of death or injury were collisions with unknown objects (46.5%), automobiles (14.1%),
utility lines (4.2%) and airplanes (1.4%); also gunshot wounds (12.7%), electrocution
(5.6%), fishing tackle (1.4%). Only 14.1% was believed due to starvation (Kerlinger
and Lein 1988).  What this suggests is that most Snowy Owls are probably in good enough
health to survive until they get the urge to return north.  All they have to do is to avoid
cars, utility lines and people with guns.

What do they eat in a world without lemmings?
Most Snowy Owls eat whatever food is locally available.  A Snowy Owl in coastal
Washington this season was seen eating a Bufflehead (duck).  Preliminary examination
of Snowy Owl pellets collected at the South Jetty of the Columbia River indicate that
those owls are catching and eating Black Rats.  An owl near Walla Walla, WA was
said to be eating voles.  Snowy Owls have been seen chasing Black-bellied Plovers
and Sanderlings.   An owl from Newport, OR was reported eating a cormorant (we
don;t know whether it caught a live one or scavenged a dead one off the beach).  An
owl pellet collected from Newport appeared to have fish bones in it, probably from
eating a bird that eats fish since there were also feathers and bird bones.

The owls I've seen just sit there.  When are they doing all this rat catching?
Snowy Owls come from the "land of the midnight sun", so it stands to reason that they
are adapted to hunt during the day (diurnal).  But during long stretches of the winter
throughout most of their range it's dark most of the day, so they must also be capable
of hunting at night (nocturnal).  They probably do a lot of their hunting at dawn and dusk
during most parts of the year.  When owls irrupt into the southern extremes of their range,
they probably adjust to local conditions and forage for food at the best time for whatever
is locally available.  And there's no point in using up energy in between meals so owls
do a lot of sitting when they're not hunting.

How old are the birds visiting us?
It is possible with some practice and good views to figure out the age of visiting owls.  It
is also possible (though a bit trickier) to figure out if they're male or female.  Females
are, on average larger than males, they also tend to have more black barring.  Young
birds also have more black barring than older ones.  A pure white (or nearly pure white)
bird is almost certainly an adult male.  Birds with mostly white heads, but extensive
barring on the wings, back and lower breast are probably young males.  In young birds,
the barring on the wing coverts is smudged together, rather than crisply black and white.
It is possible to confuse adult females with juvenile males, though adult females should have
more barring on the neck and sides of the head and lack the smudgy coverts.  Juvenile
females have lots of fairly heavy barring, including extensive barring on the back and
sides of the head.

Aren't you worried about catching a disease from handling owl pellets?
Yes, and so I wear latex gloves while collecting.  Each pellet goes in a numbered zip-lock
bag that in turn goes into another, bigger zip-lock.  When they get to my "lab" (the basement
of my house), I put on gloves again, take the pellets out of the bag and measure them.
They are then cooked in an oven for 40 mins at 325°F.  I do not do any probing or pulling
apart until after they've been cooked.  I also wash my hands thoroughly with an anti-
bacterial soap after I'm done.  After examination, the remains of each pellet are transferred
to a numbered manila envelope and stored in a safe, dry spot.

References?
GABRIELSON, I.N., and S. G. JEWETT. 1940. Birds of Oregon. Oregon State College.

JOSEPHSON, B. 1980. Aging and Sexing Snowy Owls. J. Field Ornithol. 51(2): 149-160.
On-line at: http://elibrary.unm.edu/sora/JFO/v051n02/p0149-p0160.pdf

KERLINGER, P., M.R.LEIN, and B.J.SEVICK. 1985. Distribution and population fluctuations
of wintering Snowy Owls (Nyctea scandiaca) in North America. Canadian J.  Zoo. 63:1829-1834.

KERLINGER, P., and M.R. LEIN.  1988.  Causes of mortality, fat condition, and weights
of wintering Snowy Owls.  J. Field Ornithol. 59:7-12.
On-line at: http://elibrary.unm.edu/sora/JFO/v059n01/p0007-p0012.pdf

NATIONAL AUDUBON CHRISTMAS BIRD COUNT: Historical Database
On-line at: http://www.audubon.org/bird/cbc/hr/index.html

PARMELEE, D.F.  1992.  Snowy Owl.  In The birds of North American No. 10
(A. Poole, P. Stetttenheim, and F. Gill, eds.).  The Acad. of Nat. Sci., Philadelphia, and
The Am. Ornithol. Union, Washington, DC.

SHELFORD, V.E. 1945. The relation of Snowy Owl migration to the abundance of Collared
Lemmings. Auk 62:592-596.
On-line at: http://elibrary.unm.edu/sora/Auk/v062n04/p0592-p0596.pdf

VICTOR, D. 1997. Snowy Owl reports in Cascadia, winter 1996-97.
On-line at: http://www.scn.org/earth/tweeters/snowy.html

Weir, R.D., and A.M.Lein. 1989. Snowy Owl. Hinterland who's who. Canadian
Wildl. Serv. Ottawa.
On-line at: http://www.hww.ca/hww2.asp?pid=1&id=76&cid=7