The surge



"SLAVE of THE STATE" <[email protected]> wrote in
news:[email protected]:

<snip>
> I think the end points are the stickler. My point was simply that
> both domains should be looked at -- I think that would help in
> choosing a filter. You think not?


I think that if you ignore all the things that are known to correlate
with global temperature and are aperiodic in nature, such as ENSO, then
any filtering you do in order to try to find trends in the short-term
temperature record is the intellectual equivalent of a pud pulling party.
The current flattening out of global temperature, as I said before, is
probably because we've been in La Nina conditions for 2004-2005. El Nino
was officially declared a month or so ago, and global temperatures will
go up next year (or maybe this year, there's a lag between El Nino and
temperature but I can't remember if it's from the "official" start or
when then Southern Oscillation index reverses, which is a months ahead of
the official start). Or something like that anyway.

--
Bill Asher
 
On Feb 13, 10:08 pm, William Asher <[email protected]> wrote:
> "SLAVE of THE STATE" <[email protected]> wrote innews:[email protected]:
>
> <snip>
>
> > I think the end points are the stickler. My point was simply that
> > both domains should be looked at -- I think that would help in
> > choosing a filter. You think not?

>
> I think that if you ignore all the things that are known to correlate
> with global temperature and are aperiodic in nature, such as ENSO,...


You're stuck on aperiodic. Just because something has a spectral
"energy" density at frequencies does not make it periodic. In the
strict sense, all signals are stochastic anyway (they have spectral
energy/power distributed; explore some of the papers at the NIST
Boulder site, for example, since the signals they generate are
extraordinarily "pure").

*All* filters will allow a non-zero bandwidth of frequencies through.
You couldn't see anything otherwise. No one bothers asking a question
that would take forever to answer (unless it is rhetorical, which is
to say it isn't really a question at all).

> ...then any filtering you do in order to try to
> find trends in the short-term temperature record
> is the intellectual equivalent of a pud pulling party.


I don't think I'm suggesting an emphasis of "short-term trends" to
spite any other information. I actually rejected it already.


I was steered to this:
http://www.safenow.org/