[wxqc] Calibrating anemometers

[Victor Engel]

It was a bad idea to eliminate data items with zero speed. I did that
originally because direction is undefined for zero speed and I wasn't sure
how to handle the situation. I've since decided to do a similar report with
the zeroes included. Here are the results:

 2.34   North 121
 1.71     NNE 129
 1.84      NE 151
 1.38     ENE 262
 1.81    East 303
 2.29     ESE 169
 1.41      SE 213
 1.37     SSE 384
 1.49   South 237
 1.56     SSW 88
 1.68      SW 110
  .71     WSW 94
  .83    West 35
  .63     WNW 24
 1.80      NW 56
 1.75     NNW 112
 1.62         2501

I've finished reseating the pole, and it's quite vertical now. I'll be
interested to see what, if any, difference it makes.

On 7/9/06, Victor Engel <brillig at gmail.com> wrote:
>
> After preliminary data analysis, it seems my initial impressions may have
> been faulty. Here are the results of my analysis.
>
> In this first report, I list mean DIFSPEED (my neighbor's reported speed
> minus my reported speed) for each of the coded directions. Apparently, my
> wind readings are actually higher than his. Units are miles per hour.
>
> DIFSPEED WindDirection
>    -1.46         North
>     -.36           NNE
>     -.37            NE
>     -.33           ENE
>     -.69          East
>     -.16           ESE
>    -1.17            SE
>    -1.01           SSE
>     -.84         South
>     -.33           SSW
>     -.37            SW
>     -.91           WSW
>    -1.17          West
>      .13           WNW
>    -1.16            NW
>     -.29           NNW
>     -.70       AVERAGE
>
> On the other hand, his speeds are higher for wind gust data. I think that
> is where I got the impression his readings were higher.
>
>     1.83         North
>     1.57           NNE
>     1.55            NE
>     1.38           ENE
>     1.69          East
>     1.72           ESE
>     1.03            SE
>     1.49           SSE
>     1.47         South
>     1.49           SSW
>     1.59            SW
>      .35           WSW
>     -.04          West
>      .40           WNW
>     1.79            NW
>     1.82           NNW
>     1.47       AVERAGE
>
> Finally, here is a count of total data points used for these statistics.
>
> North 83
>   NNE 96
>    NE 119
>   ENE 176
> East 166
>   ESE 98
>    SE 147
>   SSE 270
> South 169
>   SSW 45
>    SW 46
>   WSW 34
> West 23
>   WNW 15
>    NW 43
>   NNW 99
> TOTAL 1629
>
> Something that probably skews the top chart a little bit is that I only
> selected data for which windspeed was nonzero at my station. Here is the
> procedure I used:
>
> I downloaded all weatherunderground data (CSV charts) for our two weather
> stations over the period of time my station has been active. Since the times
> are sometimes every minute and sometimes every five minutes, I took only the
> first data points on or after 5 minute boundaries. Then, I selected only
> data for which both stations had values for the timeframe.
>
> I wonder if my setup is missing wind gusts for some reason or other. If
> the Davis WIND station itself does not keep track of gusts, then gusts would
> have to be captured by software, and gaps between readings could possibly
> miss gusts.
>
> Victor
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://pond1.gladstonefamily.net:8080/pipermail/wxqc/attachments/20060709/462bb856/attachment.html