Performance of Continuous

PM2.5 Monitors at a Monitoring Site in Ottawa, Canada

Tom Dann Luc White, Alain Biron

Environment Canada, ETC, Ottawa

Environment EnvironnementCanada Canada

NESCAUM Monitoring and Assessment Committee Meeting Troy, NH April 24/25, 2007

Ottawa Monitoring Site

Study Period: May 2004 – present> 500 days of data for 1 or more continuous instruments with filter-based data

24h PM2.5 Max: 64 µg/m³ 98th Perc: 37 µg/m³ Mean: 10.0 µg/m³

Temperature Range: -27°C to 29°C Filter based samplers: R&P Partisol FRM

and R&P Partisol Sequential Dichot

Ottawa Monitoring Site

Met-One BAM 1020 May 2004

Met-One BAM 1020 (new) Nov 2006

TEOM SES (30°C) May 2004

TEOM FDMS July 2004

GRIMM 180 (PM10, PM2.5, PM1.0) Oct. 2005

Thermo SHARP 5030 June 2006

Met-One BAM 1020

low-level C14 radiation source

uses beta-ray attenuation and a filter tape system

inlet is heated but filter is at ambient (station) temperature

humidity problems in early units caused higher readings in the summer

‘smart heater’ is turned on by a relative humidity sensor

hourly average output

Linear regression results for 24h dataBAM Ottawa (Warm Seasons 2004-2006)

0 10 20 30 40 50 60

Filter Based Sampler (µg/m³)

0

10

20

30

40

50

60

BA

M 1

020

(µg/

m³)

BAM 45% BAM 35%

BAM45 = -0.9 + 1.34 * FILTERr = 0.984 n = 54BAM35 = -0.6 + 1.09 * FILTERr = 0.961 n = 152

Linear regression results for 24h dataBAM with 35% Humidity Setting Ottawa (2005-2006)

0 10 20 30 40 50

Filter Based Sampler (µg/m³)

0

10

20

30

40

50

BA

M35

(µg/

m³)

BAM35_Cold BAM35_Warm

BAM35_Cold = -1.7 + 1.02 * FILTERr = 0.983 n = 72BAM35_Warm = -0.6 + 1.09 * FILTERr = 0.961 n = 152

Replacement of BAM

Linear regression results for 24h dataTEOM-FDMS Ottawa (2004-2006)

0 10 20 30 40 50 60 70 80

Filter-based Sampler (µg/m³)

0

10

20

30

40

50

60

70

80

TEO

M-F

DM

S (µ

g/m

³)

FDMS_Cold FDMS_Warm

FDMS_Cold = -0.8 + 1.07 * FILTERr = 0.993 n = 167FDMS_Warm = -0.3 + 1.08 * FILTERr = 0.969 n = 180

Grimm 180 Multi-channel Aerosol Spectrometer

Simultaneous measurement of PM1.0, PM2.5 and PM10Particle counts by size (31 channels) and mass determinationusing algorithmsNo size cut on inlet – moisture control using Nafion dryerTruly continuous (real-time)

Results for 24h PM2.5 data: Grimm 180 (Ottawa) Jan. 2006 – Dec. 2006

0 5 10 15 20 25 30 35 40

Filter-based Sampler (µg/m³)

0

5

10

15

20

25

30

35

40

GR

IMM

(µg/

m³)

GRIMM_Cold Season GRIMM_Warm Season

GRIMM_Cold = 0.5 + 1.05 * FILTERr = 0.964 n = 47GRIMM_Warm = 0.03 + 1.07 * FILTERr = 0.941 n = 60

Grimm 180 Results – Ottawa Summer 2006

05

101520253035

Conc

. (µg

/m³)

GrimmDichot

Humidity Effects on Grimm 180 (Ottawa)?

-10 -5 0 5 10 15 20 25

Dew Point Temp. (°C)

-6

-4

-2

0

2

4

6

8

GR

IMM

min

us F

ILTE

R (µ

g/m

³)

24h DataApril - September 2006

Results for 24h Coarse data: Grimm 180 (Ottawa) Jan. 2006 – Mar. 2007

0 5 10 15 20 25 30

Dichot Coarse (µg/m³)

0

5

10

15

20

25

30

GR

IMM

PM

10 -

PM

2.5 (µ

g/m

³)

GRIMM Coarse Before RecalGRIMMC = 0.8 + 0.24 * DICH r = 0.711

GRIMM Coarse After RecalGRIMMC = -1.0 + 0.82 * DICH r = 0.977

Results for 24h PM10 data: Grimm 180 (Ottawa) Jan. 2006 – Mar. 2007

0 5 10 15 20 25 30 35 40 45

Dichot PM10 (µg/m³)

0

5

10

15

20

25

30

35

40

45

GR

IM P

M10

(µg/

m³)

GRIMM PM10 Before RecalGRIMM10 = -0.4 + 0.85 * DICH r = 0.947

GRIMM PM 10 After RecalGRIMM10 = -0.2 + 0.93 * DICH r = 0.928

Grimm 180 Results – Ottawa January 29-31, 2007

01020304050607080

Conc

. (µg

/m³)

FINECOARSE

Thermo SHARP 5030 Synchronized Hybrid Ambient Real-time Particulate Monitor

Truly continuous (real-time) Combination nephelometer

& beta-ray attenuation Light scattering photometer

is continuously calibrated by beta attenuation mass sensor

Intelligent Moisture Reduction (IMR) System heating the inlet tube; threshold is set at 40%

Linear regression results for 24h data: Sharp 5030 (Ottawa Jun. – Dec. 2006)

0 5 10 15 20 25 30 35 40 45

Filter-based Sampler (µg/m³)

0

5

10

15

20

25

30

35

40

45

Ther

mo

Sha

rp (µ

g/m

³)

TSHARP = 1.20+1.00*FILTERr = 0.958 n = 68

Performance During Air Mass Change (Ottawa September 2006)

10:00 14:00 18:00 22:00 02:00 06:00 10:00 14:00 18:00 22:00 02:00 06:00 10:00

Hour of Day

-6

-4

-2

0

2

4

6

8

10

12

14

16

18

20

22

24

Con

cent

ratio

n (µ

g/m

³)

TEOM SES TEOM FDMS GRIMM THERMO SHARP

September 27 - 29, 2006

Cost Comparisons for Continuous PM2.5 Instruments (Canadian$):

Met-One BAM 1020$25,000

R&P 1400AB TEOM$25,400

R&P SES Kit $ 4,750

R&P 8500 FDMS Kit$10,700

TEOM-FDMS$36,100

GRIMM 180 (PM10, PM2.5, PM1.0)$31,000

Thermo SHARP 5030$26,500

2006 Performance Based on EPA ARM Template

Instrument N Slope Intercept

r

TEOM-SES 133 0.80 0.04 0.940TEOM-FDMS

126 1.06 -0.43 0.961

BAM 126 1.01 0.34 0.942GRIMM 107 1.06 0.27 0.953SHARP 68 1.00 1.20 0.958

PM Workshop – Ottawa April 2007

Presentations and Summary:

http://www.etc-cte.ec.gc.ca/NAPSAnnualRawData/Password: ozoneFolder: PMWORKSHOP2007