[wxqc] Re: Weather data quality report terms explained

Tue Apr 5 01:37:25 EDT 2005

Hopefully the formatting of this email will come through, if not set to
Courier Font:

Explanation of the QCMS Message

   DATE     UTC      SITE  *   ALT    * POT TEMP  *  DEW PNT *    DD    *
FF
                           *   (MB)   *  (DEG F)  *  (DEG F) *   (DEG)  *
(KNT)
03-APR-2005 2333    CW3462 *1012.(0.2)*  64( 1.8) *30.2(  -6)*   0(-55.)*
0(2.07)
03-APR-2005 Smry    CW3462 * -0.2(0.6)*-0.6(1.79)
*-3.7(1.84)*6.46(84.5)*-1.4(1.11)
03-APR-2005 Week    CW3462 *0.95(0.75)* 0.94(1.88)*-1
(2.28)*-5.6(78.6)*0.98(1.62)

You should probably just look at second and third rows, the "Smry" (summary)
and "Week" (7 days average) rows.

The acronyms at the top:

1.  ALT = Altimeter, station pressure reduced to sea level so you can
compare your pressure with other stations not at the same elevation,
pressure is shown in millibars (MB) format.

Week:  0.95(0.75), adding +0.95 millibar (+00.03 inches of mercury) to your
mean observation is needed to have exactly the mean (altimeter) pressure
indicated by the analysis.  The (0.75) is the variability (standard
deviation) of your observations relative to the mean pressure, + or minus
0.75 millibars.  When the mean error is much greater (x2) than the standard
deviation, then it is probably time for a calibration (you are doing well).

2.  POT TEMP = Potential Temperature;this is your temperature, also reduced
to sea level.  That is, if you filled a balloon at your house with "ambient"
air and drove as fast as you could to the beach, and then stuck a
temperature sensor in the balloon (now deflated somewhat), you would measure
the potential temperature. As you descend elevation, temperatures increase
due to increasing pressure (why the balloon appears to deflate).  The rule
is dry air cools 5.4F degrees every 1,000 feet you descend.  So, if your
temperature is 60.6F at your location, the (potential) temperature at the
beach will be 65.0F (nice that  you are at exactly 1,000 feet elevation!).

Week:  0.94(1.88), adding +0.94 F degrees from your average temperature will
make your average (potential) temperature the same as the surrounding
temperatures.  The (1.88) is the variability (standard deviation) of your
observations relative to the mean temperature, + or minus 1.88 F degrees.
When the mean error is much greater (x2) than the standard deviation, then
it is probably time for a calibration (you are doing well).

3.  DEW PT = Dew Point, the temperature needed to have 100% relative
humidity

Week:  -1.0 (2.28), subtracting 1.0 F degrees from your average dew point
temperature will make your average dew point temperature the same as the
surrounding average dew point temperatures.  The (2.28) is the variability
(standard deviation) of your observations relative to the mean temperature,
+ or minus 2.28 F degrees.  When the mean error is much greater (x2) than
the standard deviation, then it is probably time for a calibration (you are
doing well).

4.  DD = Wind Direction, in 360 degrees

Week:  -5.6(78.6), subtracting 5.6 F degrees from your average wind
direction will make your average wind direction the same as the surrounding
average wind direction.  The (78.6) is the variability (standard deviation)
of your observations relative to the mean temperature, + or minus 78.6 F
degrees.  When the mean error is much greater (x2) than the standard
deviation, then it is probably time for a calibration (you are doing well).

5.  FF = Wind Speed, in nautical miles per hour (KTS), 1.00 KTS = 1.15 mph

Week:  0.98(1.62), adding +0.98 knots to your average wind speed will make
your average wind speed the same as the surrounding average wind direction.
The (1.62) is the variability (standard deviation) of your observations
relative to the mean wind speed, + or minus 1.62 knots.  When the mean error
is much greater (x2) than the standard deviation (the number inside the
parentheses), then it is probably time for a calibration (you are doing
well).

-----Original Message-----
From: wxqc-bounces at lists.gladstonefamily.net
[mailto:wxqc-bounces at lists.gladstonefamily.net]On Behalf Of Alan Batie
Sent: Monday, April 04, 2005 10:39 PM
To: wxqc at lists.gladstonefamily.net
Subject: [wxqc] Re: Weather data quality report for 2005-04-04

Weather Quality wrote:

>    DATE     UTC      SITE  *   ALT    * POT TEMP *  DEW PNT *    DD    *
FF
>                            *   (MB)   *  (DEG F) *  (DEG F) *   (DEG)  *
(KNT)
> 04-APR-2005 Errs    CW2408 *          *   1/85   *          *   0/85   *
0/85
>
> Note that times are in UTC. The values displayed are  'Observed (error)'
> The error value is 'analysis - observed'.

Could someone fix this script so the description matches the actual
output: there is nothing in a format of "xxx (yyy)", nor "xxx - yyy"
(though I think that is an arithmetic statement, not a format).  If I
remember right, what this is really saying is that there was 1 error out
of 85 reports in the Pot Temp today and no other errors?

Actually, as I look through several of these, how about this to replace
that first paragraph:

Explanation:

Date: Date of the report.
UTC:  Universal Coordinated Time (Greenwich Mean Time) of the
       observation.  See below if the value is a word instead of
       a time - these entries are special entries with summary
       information.
Site: Station id of the reporting weather station.
Alt:  Barometer data in millibars.
Pot Temp: Temperature in degrees Fahrenheit
Dew Pnt: Dew point data in degrees Fahrenheit
DD:   Wind direction in degrees (0=North, 90=East, 180=South, 270=West)
FF:   Wind speed in knots

For normal timestamped entries, the values are in the format of
"Observed value(estimated error value)".  The estimated error value is
computed by subtracting the reported value from a value determined from
statistical analysis of surrounding stations, thus if the reported value
is higher than the analysis value, the error value will be negative.

If the value in the UTC column is a word, then it is a summary report:

Smry: statistical summary for the day
Week: statistical summary for the week
Mnth: statistical summary for the month

In these cases, the report data is in the format "Mean(Standard
Deviation)" over the specified period.

Errs: An error count, in the format "# errors/#reports received"
       If this occurs by itself, it is a report for the date listed.
       If it occurs with a statistical summary report, then it is the
       count of errors for the specified summary period.

Graphs:
...

Also, in the last report, the Graphs link didn't have the site value, fwiw..