The Value of On-site Comparisons During WCC Audits for Methane, - - PowerPoint PPT Presentation

Materials Sci ence & Technolog y

The Value of On-site Comparisons During WCC Audits for Methane, Carbon Dioxide and Carbon Monoxide

Christoph Zellweger1, Juha Hatakka2, Martin Steinbacher1, and Brigitte Buchmann1

1Empa, Laboratory for Air Pollution/Environmental Technology, Duebendorf, Switzerland 2FMI, Finnish Meteorological Institute, Helsinki, Finland

ESRL Global Monitoring Annual Conference, Boulder CO, USA

2

World Calibration Centre WCC-Empa

• Established 1996 for Surface Ozone
• 1997: Carbon Monoxide
• 2000: Methane
• 2007: Collaboration with WCC-N2

O

• 2010: Carbon Dioxide

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

• Since 2011, parallel

measurements with traveling instrument (CO, CH4 , CO2 )

3

History of parallel measurements during audits

• GGMT Meeting Jena 2009: It was recognized that audits using

travelling standards provide only limited information in some cases, and parallel measurements with a travelling instrument would be desirable.

• GGMT Meeting Wellington 2011:

WCC-Empa showed first results from a comparison made at Cape Point, and the following recommendation was made: ‘The World Calibration Centre for CO2 , CH4 , and CO (EMPA) has demonstrated the benefits of using a travelling instrument for GAW station audits. It is very desirable that the air intake is included in the testing process. This practice is encouraged whenever possible’.

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

4

What have we learnt from the first comparison?

• Parallel measurements using a

completely independent setup (inlet, instrument, calibration) provide very valuable additional information on the performance of a measurement system.

• It is of utmost importance to use an

independent inlet system, but addtional measurements with the travelling instrument using the station inlet system provide further useful information.

• The Picarro G2401 fully suitable as

a travelling instrument for on-site audits (CO2 and CH4 ), but improvement of water vapor correction is needed (CO).

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

4

380 390 400 410 CO2 [ppm]

WCC-Empa (Picarro G2401) CPT (URAS-4)

• 10
• 5

5 10 CPT - WCC [ppm] 11-03-17 11-04-06 11-04-26 11-05-16 ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.7 0.8 0.9 1.0 1.1 1.2 1.3 H2O mixing ratio (%) (CO)wet/(CO)dry

° Dilution only Experimental data Fit

° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° °

CO (ppb) 56 115 172

Picarro G2401

50 100 150 200

• 5

5 10 15 20 <WMO-2000> (ppb) [CO - <WMO-2000>] (ppb)

PeakPerformer 1 #084 12-04-17 09:43:30) to 12-04-18 10:58

5

Audit Pallas

Audit by WCC-Empa (April 2012)

• Station is equipped with analyzers of the newest generation.
• Agreement between PAL and WCC-Empa was good.
• An audit using travelling standards was made. Results see

below.

• In addition, the audit includes parallel measurements of CO,

CO2 and CH4 which are currently still ongoing.

WCC-Empa – GAW-CH Meeting 2012-05-16

100 200 300 400

• 0.4
• 0.2

0.0 0.2 0.4 <WMO-X2007> (ppm) CO2 - <WMO-X2007> (ppm)

Picarro G2401 #2012 12-04-18 07:04:00) to 12-04-18 14:01

CO2 Picarro

50 100 150 200

• 4
• 2

2 4 <WMO-2000> (ppb) [CO - <WMO-2000>] (ppb)

Picarro G2401 CFKADS-2018 12-04-18 06:58:30) to 12-04-18 13:55

CO Picarro CO PP1

1700 1800 1900 2000 2100

• 4
• 2

2 4 <NOAA04> (ppb) [CH4 - <NOAA04>] (ppb)

Picarro G2401 CFKADS-2018 12-04-18 07:04:00) to 12-04-18 14:01

CH4 O3 Analyzer

20 40 60 80 100 120

• 10
• 5

5 10 SRP (ppb) [OA - SRP] (ppb)

Sufficient Good Good Sufficient TEI49i #619917500 Unbiased ozone = OA - 0.11 (ppb) / 0.983 340 350 360 370 380 390 400

• 0.2
• 0.1

0.0 0.1 0.2 0.3 0.4 <WMO-X2007> (ppm) CO2 - <WMO-X2007> (ppm)

Licor LI7000 12-04-18 17:57:00) to 12-04-18 17:57

CO2 Licor

6

Parallel CO2 / CH4 / CO Measurements at PAL

• Picarro G2401 was used as a travelling instrument.
• Comparison ongoing since 20. April, planned to continue till June.
• A completely independent inlet line was used; same air intake

location as for the PAL instruments.

• In addition, automatic measurements using the same inlet as PAL

are made every 30 h during a period of 10 h.

• No sample drying was used for the Picarro instrument; a water

vapor correction was applied for CO2 and CH4 .

• Two working and one target tank are measured every 40 h.

Installation of separate inlet line at Pallas PAL CO instrument (RGD) PAL CO2 instrument (NDIR) PAL CO2 /CH4 /CO instrument (CRDS)

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

7

Comparison between the two Picarro G2401

Pallas:

• Picarro G2401 #2018 instrument.
• Pallas air inlet system was used.
• The Instrument was regularly calibrated using a target tank.
• A Nafion drier was used for sample air drying.

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

WCC-Empa:

• Picarro G2401 #2001 travelling instrument.
• A completely independent inlet line was used;

same air intake location as for the PAL

• instruments. In addition, automatic measurements

using the same inlet as PAL are made every 30 h.

• No sample drying was used for the Picarro

instrument; a water vapor correction was applied for CO2 and CH4 .

0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.96 0.97 0.98 0.99 1.00 water vapor mixing ratio (%) (CO2)wet/(CO2)dry Dilution only Experimental data Fit

(CO2)wet/(CO2)dry = 1 + -0.01198 * (Hrep) + -0.00026 * (Hrep)^2

0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.965 0.970 0.975 0.980 0.985 0.990 0.995 1.000 water vapor mixing ratio (%) (CH4)wet/(CH4)dry Dilution only Experimental data Fit

(CH4)wet/(CH4)dry = 1 + -0.01022 * (Hrep) + -0.00012 * (Hrep)^2

8

Parallel CO2 Measurements – Picarro G2401

• The WCC-Empa Picarro

instrument is running very stable according to the regular measurements of the working tanks.

• The overall agreement

between the two instruments is extremely good.

• No obvious difference

between the different air inlets.

• WCC-Empa measurements were made without sample drying, and a

correction was applied to the CO2 data based on an experiment in which the water vapor influence was determined.

• PAL measurements were made using a Nafion drier.

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

399 400 401 402 403 CO2 [ppm] PAL Picarro G2401 WCC Picarro (WCC inlet) WCC Picarro (PAL inlet)

• 0.3
• 0.2
• 0.1

0.0 0.1 CO2(PAL) - CO2(WCC) [ppm] PAL - WCC (WCC inlet) PAL - WCC (PAL inlet) 12-04-25 12-04-30 12-05-05 Bias (ppm) Frequency (counts)

• 0.5
• 0.3
• 0.1

0.1 500 1500 2500

Bias (ppm) mean median st.dev

• 0.02
• 0.02

0.03

399 400 401 402 403 CO2 [ppm] PAL Picarro G2401 WCC Picarro (WCC inlet) WCC Picarro (PAL inlet)

• 0.3
• 0.2
• 0.1

0.0 0.1 CO2(PAL) - CO2(WCC) [ppm] PAL - WCC (WCC inlet) PAL - WCC (PAL inlet) 12-04-25 12-04-30 12-05-05 399 400 401 402 403 CO2 [ppm] PAL Picarro G2401 WCC Picarro (WCC inlet) WCC Picarro (PAL inlet)

• 0.5

0.0 0.5 CO2(PAL) - CO2(WCC) [ppm] PAL - WCC (WCC inlet) PAL - WCC (PAL inlet) 12-04-25 12-04-30 12-05-05 9

Comparison CO2 NOAA-Flasks – FMI data

• FMI data ± 5 min average..
• Time matching is very

important.

• Small differences of the

temporal coverage between flask samples and selected continuous data might explain part of the observed bias.

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

Bias (ppm) Frequency (counts)

• 0.5
• 0.3
• 0.1

0.1 500 1500 2500

Bias (ppm) mean median st.dev

• 0.02
• 0.02

0.03

• 0.5

0.0 0.5 CO2(FMI-NOAA) [ppm] 09-01-01 10-01-01 11-01-01 12-01-01

Bias (ppm) Frequency (counts)

• 1.0
• 0.5

0.0 0.5 1.0 5 15 25 35

Bias (ppm) mean median st.dev

• 0.07
• 0.05

0.22

FMI – WCC Empa FMI – NOAA

Bias (ppm) Frequency (counts)

• 0.5
• 0.3
• 0.1

0.1 500 1000 1500

Bias (ppm) mean median st.dev

• 0.02
• 0.02

0.08

FMI – WCC Empa shifted by 5 min

10

Parallel CO2 Measurements – Licor LI7000

• The overall agreement between

the two instruments is good; an average bias of -0.05 ppm CO2 was observed based on 1-min data.

• The Licor instrument is

connected to the same air inlet as the PAL Picarro.

• The Licor data was corrected

based on the results of the audit measurements.

• WCC-Empa measurements were made

without sample drying.

• PAL Licor measurements were made

using a Nafion drier as pre-drying followed by a Mg(ClO4 )2 cartridge.

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

399 400 401 402 403 CO2 [ppm] PAL Licor 7000 WCC Picarro (WCC inlet) WCC Picarro (PAL inlet)

• 0.4

0.0 0.2 0.4 CO2(PAL) - CO2(WCC) [ppm] PAL - WCC (WCC inlet) PAL - WCC (PAL inlet) 12-04-25 12-04-30 12-05-05 Bias (ppm) Frequency (counts)

• 0.8
• 0.4

0.0 0.4 500 1000 1500

Bias (ppm) mean median st.dev

• 0.05
• 0.05

0.06

340 350 360 370 380 390 400

• 0.2
• 0.1

0.0 0.1 0.2 0.3 0.4 <WMO-X2007> (ppm) CO2 - <WMO-X2007> (ppm)

Licor LI7000 12-04-18 17:57:00) to 12-04-18 17:57

11

Parallel CH4 Measurements – Picarro G2401

• The overall agreement between

the two instruments is good; an average bias of -0.08±0.36 ppb CH4 was observed based on 1- min data.

• As for CO2

, no difference between the PAL and WCC inlet was observed.

• WCC-Empa measurements were made without sample drying, and a

correction was applied to the CH4 data based on an experiment in which the water vapor influence was determined.

• PAL measurements were made using a Nafion drier.

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

1890 1910 1930 CH4 [ppb] PAL Picarro G2401 WCC Picarro (WCC inlet) WCC Picarro (PAL inlet)

• 4
• 2

1 2 3 CH4(PAL) - CH4(WCC) [ppb] PAL - WCC (WCC inlet) PAL - WCC (PAL inlet) 12-04-25 12-04-30 12-05-05 Bias (ppb) Frequency (counts)

• 3
• 2
• 1

1 2 3 500 1500

Bias (ppb) mean median st.dev

• 0.08
• 0.09

0.36

1890 1910 1930 CH4 [ppb] PAL Picarro G2401 WCC Picarro (WCC inlet) WCC Picarro (PAL inlet)

• 4
• 2

1 2 3 CH4(PAL) - CH4(WCC) [ppb] PAL - WCC (WCC inlet) PAL - WCC (PAL inlet) 12-04-25 12-04-30 12-05-05 1890 1910 1930 CH4 [ppb] PAL Picarro G2401 WCC Picarro (WCC inlet) WCC Picarro (PAL inlet)

• 4
• 2

1 2 3 CH4(PAL) - CH4(WCC) [ppb] PAL - WCC (WCC inlet) PAL - WCC (PAL inlet) 12-04-25 12-04-30 12-05-05 12

Comparison CH4 NOAA-Flasks – FMI data

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

• FMI data ± 5 min average..
• Time matching is very important.
• However, the difference cannot

be fully explained by time matching.

• The bias was also observed

during the analysis of the WCC TS on the PAL Picarro. However, an issue with the humidification

• f the standards on the Nafion

needs further attention.

• 4
• 2

2 4 CH4(FMI-NOAA) [ppb] 09-01-01 10-01-01 11-01-01 12-01-01

Bias (ppb) Frequency (counts)

• 4
• 2

2 4 5 10 15 20

Bias (ppb) mean median st.dev 1.24 1.5 1.21

FMI – WCC Empa FMI – NOAA

Bias (ppb) Frequency (counts)

• 4
• 2

2 4 500 1500

Bias (ppb) mean median st.dev

• 0.08
• 0.1

0.49

FMI – WCC Empa shifted by 5 min

Bias (ppb) Frequency (counts)

• 3
• 2
• 1

1 2 3 500 1500

Bias (ppb) mean median st.dev

• 0.08
• 0.09

0.36

13

Parallel CO Measurements – Picarro G2401

• An average bias of 4.5±1.4 ppb

CO was observed based on 1-h data.

• As for CO2

, no difference between the PAL and WCC inlet was observed.

• WCC-Empa measurements

were made without sample drying.

• PAL measurements were made

using a Nafion drier.

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

120 130 140 150 160 170 CO [ppb]

PAL WCC with seperate inlet WCC using PAL inlet

• 4

2 4 6 8 date [yy-mm-dd] PAL.CO - WCC.CO [ppb] 12-04-25 12-04-30 12-05-05

Reason of the bias?

50 100 150 200

• 4
• 2

2 4 <WMO-2000> (ppb) [CO - <WMO-2000>] (ppb)

Picarro G2401 CFKADS-2018 12-04-18 06:58:30) to 12-04-18 13:55

CO Picarro

14

Parallel CO Measurements – PeakPerformer1

• Similar bias of 5.3±0.8 ppb CO

was observed based on 1-h data.

• Pallas Picarro G2401 and

PeakPerfomer agree well, but WCC-Empa measurements are biased.

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

120 130 140 150 160 170 CO [ppb]

PAL PP1 WCC with seperate inlet WCC using PAL inlet

• 4

2 4 6 8 date [yy-mm-dd] PAL.PP1.CO - WCC.CO [ppb] 12-04-21 12-04-22 12-04-23 12-04-24 12-04-25 12-04-26 12-04-27

Reason of the bias?

50 100 150 200

• 5

5 10 15 20 <WMO-2000> (ppb) [CO - <WMO-2000>] (ppb)

PeakPerformer 1 #084 12-04-17 09:43:30) to 12-04-18 10:58

CO PP1

15

Reason of the bias?

• Agreement of travelling standard comparison was good (Picarro), and PAL was higher

compared to WCC-Empa (PeakPerformer1).

• Unexpected instrument drift?
• Water vapor correction?

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.7 0.8 0.9 1.0 1.1 1.2 1.3 H2O mixing ratio (%) (CO)reported/(CO)dry ° Experimental data Linear fit ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° CO2 = 413 ppm

CO (ppb) 66 126 246 185

° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.7 0.8 0.9 1.0 1.1 1.2 1.3 H2O mixing ratio (%) (CO)reported/(CO)dry ° Experimental data Linear fit ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° CO2 = 413 ppm

CO (ppb) 69 128 247 188

° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.7 0.8 0.9 1.0 1.1 1.2 1.3 H2O mixing ratio (%) (CO)reported/(CO)dry ° Experimental data Quadratic fit ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° ° °

CO (ppb) 56 115 172

G2401#2001 before optimization G2401#2001 after optimization G2401#2028

16

Parallel measurements with CO analyzers…

• Picarro G2401 (CDRS).
• Aerolaser AL5001 (VURF).
• Aerodyne Mini-QCL.
• LGR-23d (ICOS-QCL).
• Good agreement for 1-h

values over a period of one week.

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012

100 200 300 400 500 CO [ppb]

CRDS VURF Mini-QCL ICOS-QCL

• 8
• 4

4 Date (yy-mm-dd) Bias to ICOS-QCL (ppb)

WMO/GAW DQO

11-09-17 11-09-18 11-09-19 11-09-20

• C. Zellweger et al., to be submitted to AMTD

17

Conclusions

Parallel measurements during audits provide …

• …

a Verification of inlet design / inlet suitability…

• …
• r they help to identify problems with a set-up
• …

include an assessment of the influence of sample drying

• …

are an independent check that includes the whole measurement set-up (inlet, instrumentation, air pretreatment, analysis, calibration, data processing).

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012 399 400 401 402 403 CO2 [ppm] PAL Picarro G2401 WCC Picarro (WCC inlet) WCC Picarro (PAL inlet)

• 0.3
• 0.2
• 0.1

0.0 0.1 CO2(PAL) - CO2(WCC) [ppm] PAL - WCC (WCC inlet) PAL - WCC (PAL inlet) 12-04-25 12-04-30 12-05-05

380 390 400 410 CO2 [ppm]

WCC-Empa (Picarro G2401) CPT (URAS-4)

• 10
• 5

5 10 CPT - WCC [ppm] 11-03-17 11-04-06 11-04-26 11-05-16

The current comparison at PAL clearly shows that …

• …

drying of the air with a Nafion dryer is not a problem …

• …

but it is also not really needed, at least for CO2 and CH4

• …

the bias that was observed for the CO comparison needs to be further investigated.

• …

such measurements provide clearly additional information which can only be partly achieved with travelling standard comparisons or round robins.

• WCC-Empa will continue using travelling instruments during
• n-site audits whenever it is feasible.
• Next comparisons are planned to take place at Zeppelin (Ny

Ålesund) and Cape Verde GAW stations later this year

18

Thank you!

Acknowledgments

• PICARRO Inc. for lending one of their G2401 instruments.
• Pallas team for helping us with the comparison exercise.
• Financial support from MeteoSwiss and Synergy / Collaboration with NABEL (Empa and FOEN)
• Empa co-workers for sharing their experience and data

WCC-Empa – NOAA ESRL Global Monitoring Annual Conference, May 2012