I found similar things Liz when I imported 2yrs of data from WU - and found 
that many of the negative values were a result of either -- or 9999 values 
in the WU data.  I initially did a quick QC by looking for negative winds, 
humidity and barometer, humidity > 100, barometer > 1300, barometer < 800, 
temp > 50, temp< -20, dewpoint < -20, heatindex > 60, windchill < -30 and 
by the end of it all have got what I think is an OK set of data.  I agree 
though that there is definitely room for basic QC values to be put in the 
wee_import config file perhaps - but I believe a more generic and across 
the board 'solution' is being considered by Gary as I write

On Tuesday, 6 September 2016 23:53:42 UTC+3, Liz wrote:
>
> On Mon, 5 Sep 2016 19:08:45 -0700 (PDT) 
> gjr80 <[email protected] <javascript:>> wrote: 
>
> > > I'm not worried about the bad data that may  or will arrive - I can 
> > > check the database for that later 
> > >   
> > 
> > As long as you are aware. --calc-missing can cause one bad field to 
> > propagate to others. 
>
> A quick review has shown some strange data arrive. 
> One day has over 1200mm of rain in the import (yes, we did flood, but I 
> thought it was about a tenth of that volume arrived). 
> The table that can be viewed on WU shows 56.4mm that day. 
>
> Other things of "interest" are an average wind speed one day which is 
> negative. I was aware that some temperatures and wind speeds were 
> higher than they should have been, I think because the old weather 
> station sent bad data. 
>
> I didn't use --calc-missing. 
>
> There is scope for quick QC on the incoming data from WU. I'll tidy 
> mine up by deleting ridiculous values. 
>
> Liz 
>

-- 
You received this message because you are subscribed to the Google Groups 
"weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to