This year there is 1) warm weather and 2) little snow, so that seems
to suggest "early" again. But it occurs to me that other factors
could be computed into a prediction model; like, a correlation of
past break-out times with temps and snowfall predictors of certain
periods and over-all; correlation of ice thickness and type of ice
grains; correlation of gravity and phases of moon; etc..
Someone able to model the predictors for computer analysis with
number crunching capability could sell me results in the form of a
handicapper's booklet or floppy so I could do permutations on my
own. All I want is a fraction of the pot, like $20-30K.
bookburn
So it's the water. Must be water temp and volume of run off are
important; I suppose that reflects ambient temperatures; and then
there is that thing about sunlight reflecting in the ice to cause
holes. bookburn
Little snow during the winter, means no insulation... so the ice
gets thicker for any given number of cold days.
Little snow melt (either because it's too cool in the spring or
because there is no snow), means no water to push the ice out.
I think you are right about spring weather having more to do
with it than ice thickness. The really weird breakups I've seen
have been when there is little snow cover during the cold parts
of the winter, causing really thick ice, and then a cool spring
that allows the spring melt to slowly run off without ever
getting enough water at once to raise the river high enough to
crack the ice. Then the ice is left there to just rot away
slowly.
Of course, for most folks living on the banks... that's just a
*great* breakup! The bad ones are when there is thick ice and
lots of snow, and it gets _real_ warm and melts it all in a few
days. All that water comes running down the rivers, cracks up
all that thick ice just enough to get it to move. And it jams
and floods the Hell out of all the river villages.
There ain't nothing like visiting your neighbors in a boat and
tying off to somebody's (maybe your own) front porch with three
or four feet of water under the keel. And of course the prelude
to that is just nerve wracking... that is seeing bits and
pieces of houses from up the river in the ice directly out in
front of your village when the ice is jamming down stream a few
miles and the water is rising. (We make jokes now and then
about Alaskan's getting cabin fever and acting weird, but to
really see people's nerves jangled just go visit any river
village about the time breakup comes along and it looks like a
flood.)
--
Floyd L. Davidson <http://www.ptialaska.net/~floyd>
Ukpeagvik (Barrow, Alaska) fl...@barrow.com
The most significant thing is the amount of sunlight and the
ambient temperature. But not at a place like Nenana! Its back
up in the hills and out in the lakes and on the tundra. The
places that feed water to the river. I'm not real familiar with
the Nenana River, but it's a good bet that the lower half of the
river has significantly less effect on the water level in Nenana
that does the upper half.
With a small river like the Nenana that is still relatively
close by though, compared to the Kuskokwim or Tanana, where the
big effects might be 3-400 miles away from the worst
floods... or heaven forbid the Yukon where it can be twice or
more that distance.
You can download statistics programs. (I found a good one for Mac
when I took stats in college.) You could try plugging in as much
data as you could find and see what happens. Actually, that sounds
like something that would be fun to mess with : )
Jan
> > Someone able to model the predictors for computer analysis with
> > number crunching capability could sell me results in the form of a
> > handicapper's booklet or floppy so I could do permutations on my
> > own. All I want is a fraction of the pot, like $20-30K.
> >
> > bookburn
Jan Flora:
> You can download statistics programs. (I found a good one for Mac
> when I took stats in college.) You could try plugging in as much
> data as you could find and see what happens. Actually, that sounds
> like something that would be fun to mess with : )
I may very well be F.O.S. so take this for what it's worth. And none of
what I think has anything to do with whether or not it would be *fun* to
mess with this just whether it would be effective 8-) But.....
Long time ago in a statistics and/or modeling class we talked about this,
maybe someone even researched it; I don't know. I believe you'll find that
people have indeed put some effort into this both as an academic exercise
and less formally for betting. I remember concluding for myself at that
time that it's not at all predictable, given the factors that can be
actually be identified and measured at the time bets are placed. What
factors are important? Are those factors even measured or measurable in the
huge *unpopulated* and nearly inaccessable area drained by the Tanana? You
probably know that it includes a large part of the Wrangel and Alaska
Ranges, and a huger (?) area of hills and rivers north of the river. Not to
mention the influence of the Tanana downriver, the Nenana River, even Yukon
river factors affecting up river ice movement.
Maybe most importantly, weather factors that happen *after* the early April
deadline for betting probably have the most influence. Breakup, I believe
happens several weeks to a month or more after the betting deadline.
Temperatures, precipitation, etc. in those weeks can completely change the
scenario predicted by the ice depth, snow pack in the drainage, etc. that
exists in early April. Not to mention random day-to-day factors that could
affect the flow of the water in front of the village of Nenana. There's
been years when the water flowed freely down the center of the channel and
the tripod stood still on shore-fast ice... and vice versa. Ice jams just
down river at the confluence with the Nenana can affect movement for days or
weeks.
Ah, for what it's worth. I'm not a betting guy and a buy and hold investor
as well. Maybe someone's proven me wrong in the meantime, but I don't think
so 8-)
DJ
kurt
"bookburn" <book...@yahoo.com> wrote in message
news:v43l0ka...@corp.supernews.com...