Loading ...
Sorry, an error occurred while loading the content.

Re: MADIS Weather Quality Charts

Expand Messages
  • schaka1
    WUHU, thanks for the info! There s a METAR 4 miles from me. My Data is always in close agreement with it. I ve often wondered how I can be in such close
    Message 1 of 3 , Oct 13, 2006
    • 0 Attachment
      WUHU, thanks for the info! There's a METAR 4 miles from me. My Data
      is always in close agreement with it.

      I've often wondered how I can be in such close agreement with this
      METAR and show so much variability within MADIS!

      Chuck

      --- In wuhu_software_group@yahoogroups.com, "wuhu_software"
      <wuhu_software@...> wrote:
      >
      > schaka1,
      >
      > I think it means 1 degree F spread (plus or minus) about the mean
      > (average).
      >
      > My question has always been how large of an area is considered to
      be
      > in the data set used to compare other stations? Is it a fixed
      radius
      > from your location?
      >
      > If it covers a large area (as I suspect it does), a temperature
      > variability of 1 degree is actually quite good IMHO.
      >
      > It seems you would have to have very stable conditions over a wide
      > area to achieve better average errors.
      >
      > You might also see low standard deviation errors if you are a lone
      > station in a given area or there are very few other stations to
      > compare with. Of course those stations would need to closely match
      > your own in order to see lower errors around the mean.
      >
      > The bottom line is that you could have a real 1/10 of a degree
      > temperature accuracy and still have a 1 degree F standard deviation
      > in the analysis data.
      >
      >
      > --- In wuhu_software_group@yahoogroups.com, "schaka1" <blkguy@>
      > wrote:
      > >
      > >
      > > I'm noticing something interesting about my quality charts. I'm
      > > wondering if anyone else is seeing the same thing.
      > >
      > > It has specifically to do with the scatter diagrams. Without
      going
      > into
      > > a lot of statistically stuff what a scatter diagram shows is a
      > central
      > > tendency. The dots should be clustered around some value if they
      > are all
      > > shooting for the same target.
      > >
      > > On my charts most of the dots are clustered about one standard
      > deviation
      > > higher than the target. This is for all parameters except
      Barometric
      > > Pressure.
      > >
      > > Anybody have any idea why that is?
      > >
      > > http://weather.gladstonefamily.net/cgi-bin/wxqual.pl?site=C5229
      > > <http://weather.gladstonefamily.net/cgi-bin/wxqual.pl?site=C5229>
      > >
      >
    Your message has been successfully submitted and would be delivered to recipients shortly.