[ExI] i got friends in low places

[spike at rainier66.com]

...> On Behalf Of BillK via extropy-chat
...
>
>>... A new correlation is becoming compelling: the quiet Atlantic with
droughts...
>>... BillK, is Jolly Olde having a dry year please?
>
> spike
> _______________________________________________


>...Oh, yes!  Same as Europe.  Grass lawns have disappeared, all dried up.
Water restrictions and droughts declared.
Fire risk very high. Record heat temperatures hit this summer.
My suntan is the best for years!  :)


BillK

_______________________________________________


BillK, I am following all this for a reason.  If you are in a hurry, skip to
the second to last paragraph below, which is the point of this rambling
epistle.

When one knows a powerful math tool technique such as Kalman filtering, one
is compelled to find applications for it everywhere just to see how they do.
I am fooling with a Kalman filter to predict Accumulated Cyclonic Energy
(ACE)

https://en.wikipedia.org/wiki/Accumulated_cyclone_energy

but while I was doing that... I realized there is something else hiding in
plain sight which is even more important.

It is cool to predict ACE for insurance purposes (the insurance companies
need to estimate how much to charge in order to make money (and hurricanes
are exciting (so they sell a lotta headlines and ad clicks))) but what about
predicting drought?  Those who have lived in Florida already know that if
one goes a whole season without a hurricane anywhere near (within a coupla
hundred miles for instance) then that will be a really dry fall and winter,
which has consequences.  

For Floridians, a direct hit by a hurricane is a really bad thing, but in
general a hurricane nearby is a good thing.  Any hurricane which doesn't
wreck your town is good, just as in California any small earthquake which
doesn't wreck stuff is good (because it relieves strain on the fault and
compels proles to bolt their shelves.)  A hurricane in Florida brings
much-needed rain because they don't have rivers to bring them water (it's
too flat there.)  They don't even have major pipelines in general.  They are
dependent on ground water, which is replenished by rainwater, which is
dependent on storms, such as the one BillW is hogging now.  Florida needs
BillW's storm.  BillW needs Florida to take his storm off his wet hands.

But I digress.  I am fooling with a Kalman filter which puts our available
observables (Atlantic surface temperatures) into my covariance matrix to
predict European droughts.  Reasoning: although it doesn't directly wreck
things and make for great news clips, a long drought is more destructive
than a hurricane, Hell I like hurricanes.  But I don't like droughts.  A
good predictor of drought is more valuable than that lame accumulated
cyclonic energy metric.

We are now 7 days from moving 2022 into second place for the longest span
between named storms in the Atlantic.  The current outlook shows two
depressions, each with an estimated less than 20% chance of forming a
tropical storm.  My calcs suggest that this means we have less than about
36% chance of a named storm in the next 5 days, which means more than 64%
chance we get to move into second place for most pacific Atlantic, in a year
when sea surface temperatures are high.  We should be seeing a wet stormy
year according to prevailing standard theory, but we are seeing a dry quiet
year.  Think about that.  

My Kalman filter is having a great time.  Or rather it would be enjoying
itself, if matrix calculus could have a great time.

spike