Technical Audit Methods and Results for Anemometers Used in the Missouri Ambient Air Monitoring Networks

1
Technical Audit Methods and Results for
Anemometers Used in the Missouri Ambient Air
Monitoring Networks
Stephen M. Hall Missouri Department of Natural
Resources Air Quality Assurance Unit
2
Introduction

• Wind speed and wind direction meteorological data
  are monitored at many ambient air monitoring
  sites in Missouri. Of the two quantities, wind
  direction has shown to be the most inaccurate.

This presentation identifies an interpretation of
the EPA wind direction data accuracy criteria and
presents a method that was used to collect
independent audit data. This data was analyzed
to determine anemometer orientation accuracy.
3
Outline

• EPA Wind Direction Accuracy Criteria (an
  interpretation)

• Audit Method Development

• Audit Equipment Utilized

• Audit Results

• Audit Equipment Precision Accuracy

• Closing Points

• Recommendations

4
EPA Wind Direction Criteria

• Wind Direction Accuracy

?5 Degrees total System error (orientation
error sensor index error)
? 3 degrees relative to the Anemometers (sensor)
mount or index (? 5 Degrees absolute error for
installed systems)
5
Interpretation of EPA Wind Direction Criteria for
Technical Auditing

• Absolute Error of Anemometers (sensor)
  alignment notch (index) from True North
• ? 2 degrees acceptable for orientation error.

• Allows vane angle response errors to be ? 3
  degrees and the total system error will remain
  within 5 degrees.

• The trick is to determine what is required to
  measure and verify that alignments are within 2
  degrees of True North.

6
Combined Index and Orientation Error - Graphical
Description
Orientation Error (/-)
Anemometer Alignment Notch (index) (its physical
orientation)
True North
Vane Angle Response error /- (relative to Notch)
7
Audit Method for Orientation

• Measure location of alignment notch with a
  precision compass (resolution /- 0.5 degrees,
  Accuracy 0.25 degrees).

• Determine True North by compensating for the
  sites magnetic declination.

• Find the difference between True North and the
  anemometers orientation. This difference is the
  absolute error from True North which is
  independent of the Anemometers response relative
  to its alignment notch.

8
Variety of Anemometer Installations
21- RM Young Model 05103 or 05305-(AQ model)
(integral Vane Prop) (there were only 2 of the
21 that were the AQ model)

• 5- Mesa Systems Co. (Ultrasonic)

1- Climatronics (separate cup and Vane)

• Due to the overwhelming majority of the RM-young
  Anemometers, we developed the audit protocol
  specifically for these units.

9
Variety of Tower Installations

• 10 meter towers (fold down or telescoping)

• 6 meter towers (pole installations on top of
  sampling shelters)

Note Only about 10 of the 27 sites have
anemometers mounted on towers 10 meters above
ground level.
10
Typical 10 Meter Tower Installation
11
Typical 6 meter tower installation
12
Orientation Audit Equipment

• Due to the variety of Anemometer installations a
  flexible audit orientation apparatus was
  constructed

13
Orientation Audit Equipment
14
Orientation Equipment
15
Orientation Equipment
16
Compass Precision Alignment Assembly
17
Orientation Audit Results

• The following orientation audit data represent
  audit results from all 25 of the Networks unique
  metrology sites which consist of the RM-Young
  installation.

• The sign of the difference was arbitrarily
  defined as follows

() measured orientation leads True North
() measured orientation lags True North
18
(No Transcript)
19
Anemometer Vane Angle Response Error Relative to
Alignment Notch (index)

• Determines the error of the anemometers degree
  response relative to the alignment notch (index)

20
Anemometer Installed for an Audit
21
Alignment Notch Interface
Alignment Notch Interfaced with bench notch
(index)
Align notch here, and you have made a 60 degree
mistake! -Be CAREFUL-
22
Vane Angle Response Audit Results

• 25 of the networks sites consisting of the
  RM-Young models were audited over a 3 year
  period.
• The audit data represent all paired differences
  (550 total differences) over the following
  discreet audit points
• 0, 30, 60, 90, 150, 180, 230, 270, 330, and 350
  degrees.

23
(No Transcript)
24
(No Transcript)
25
Total System Error As Found

• The following histogram shows the combined effect
  of the orientation and vane angle response error
  with the anemometer in its as found condition.

26
(No Transcript)
27
Total System Error Adjusted

• The following histogram shows the combined effect
  of the orientation and vane angle response error
  on Total System Error AFTER the anemometers
  alignment notch was aligned to True North.

28
(No Transcript)
29
(No Transcript)
30
Audit Equipment Precision Accuracy

• The accuracy of the audit equipment is determined
  by the manufactures published specifications.

• We gathered data to check the repeatability of
  the following Audit equipment

• Vane angle response bench
• Orientation apparatus

31
Vane Angle Response Bench

• Blind repeatability test- Verified the precision
  of our vane angle response bench with two setups
  and the same anemometer.

• Maximum error for any repeated setting was within
  0.24 degrees.

32
(No Transcript)
33
(No Transcript)
34
Orientation Equipment

• Anecdotal evidence suggests repeatability will be
  well within one degree.

35
Comparing Two Sources of Declination Data

• Aeronautical Chart , US DOT FAA April 18, 2002

NOAA Site http//www.ngdc.noaa.gov/seg/geomag/jsp
/IGRF.jsp for April 18, 2002
36
(No Transcript)
37
Closing Points

• Orientation error is the greatest contributor to
  the inaccuracy of wind metrology data.

• Vane angle response errors are the greatest
  contributors to anemometer performance audit
  failure.

• Infrequent quality control checks correlate with
  anemometer performance audit failure.

38
Recommendations

• Magnetic Interference must be carefully
  considered when developing and implementing audit
  protocol for the orientation accuracy. Keep in
  mind that you may need to address this issue on a
  site specific basis.

• When developing vane angle response criteria,
  establish audit points that monitor anemometer
  performance at both limits of the of
  potentiometer Dead Zone.

39
Recommendations

• Monitoring agencies should establish routine
  quality control checks of the anemometers bench
  performance at least once every six months.
• Agencies that have independent auditing could
  reduce quality control checks to once per year
  with audits performed once per year and,
  preferably, 6 months out of phase with the QC
  check.

40
Recommendations

• Chose a particular source for the networks
  declination estimate and use it consistently when
  performing orientations.