[wxqc] Analysis Anomalies varying between sites

[Mike Barth]

Victor,

In this case, the C6155 analysis did in fact use a station that was much
lower than your data and most of the other stations used in the analysis,
and this station wasn't used in the analysis for KATT.

One thing to remember is that there is a lot of data that can be used in
the analyses that is not available for viewing on these plots.  The analysis
values should only be used to identify large biases.

Mike

On Sun, 24 Sep 2006, Victor Engel wrote:

> I have linked here an analysis for my station and KATT, the closest station
> to mine. Notice the weird jump in humidity near the left end of the graphs.
>
> http://weather.gladstonefamily.net/qchart/C6155?date=20060924&addnl=KATT&Add+to+charts=Add+to+charts&.cgifields=addnl
>
> OK. Now let's look at the analysis for KATT with my station added. This
> should be essentially the same graph but with a different analysis.
>
> http://weather.gladstonefamily.net/qchart/KATT?date=20060924&addnl=C6155&Add+to+charts=Add+to+charts&.cgifields=addnl
>
> Notice how close the analysis tracks my station? I guess that's no surprise,
> really, since, if my station is used in the analysis for KATT, they should
> have a good correlation. So then what causes the anomaly on my analysis?
> Presumably, it would involve a station used in my analysis and not one for
> KATT. But adding all the stations with checkboxes, I don't see any station
> that correlates to the analysis.
>
> Here is a link showing all the sites with checkboxes on my analysis that are
> not on the KATT analysis.
> http://weather.gladstonefamily.net/qchart/C6155?date=20060924&addnl=KATT&addnl=AR839&addnl=AR384&addnl=C6261&Add+to+charts=Add+to+charts&.cgifields=addnl
>
> The only station correlating the the blip is KATT, but the amplitude is much
> greater for the analysis than for KATT. Does the analysis think there is
> some weather feature that is closer to me than to KATT because none of
> KATT's neighbors show it? Or is the analysis based upon stations not present
> on either list? Or is there simply a problem with the analysis, or something
> else?
>