1

LaboratoryEvaluation ofTwo OpticalInstruments forReal-TimeParticulateMonitoring ofSmoke

LaboratoryEvaluation ofTwo OpticalInstruments forReal-TimeParticulateMonitoring ofSmoke

United StatesDepartment ofAgriculture

Forest Service

Technology &DevelopmentProgram

2500 WatershedFebruary 19999925-2806-MTDC

2

The U.S. Department of Agriculture (USDA) prohibitsdiscrimination in all its programs and activities on thebasis of race, color, national origin, gender, religion,age, disability, political beliefs, sexual orientation, andmarital or family status. (Not all prohibited bases applyto all programs.) Persons with disabilities who requirealternative means for communication of programinformation (Braille, large print, audiotape, etc.) shouldcontact USDA’s TARGET Center at 202-720-2600(voice and TDD).

To file a complaint of discrimination, write USDA,Director, Office of Civil Rights, Room 326-W, WhittenBuilding, 14th and Independence Avenue, SW,Washington, DC 20250-9410 or call (202) 720-5964(voice or TDD). USDA is an equal opportunity providerand employer.

LaboratoryEvaluation ofTwo OpticalInstrumentsfor Real-TimeParticulateMonitoring ofSmoke

LaboratoryEvaluation ofTwo OpticalInstrumentsfor Real-TimeParticulateMonitoring ofSmoke

Andy TrentProject Leader

Harold ThistleProgram Leader

Rich FisherAir Resource Management, Washington Office

Ronald BabbittElectronics Engineer, Fire Sciences Laboratory

Andria Holland-SearsHydrologist, White River National Forest

USDA Forest ServiceTechnology and Development ProgramMissoula, Montana

8E82F54—Evaluate Two Nephelometers

February 1999

3

Table of Contents

Introduction _______________________________________________________ 1

Instruments _______________________________________________________ 2

Methods __________________________________________________________ 4

Results and Discussion ___________________________________________ 5

Conclusions _____________________________________________________ 10

Additional Tests __________________________________________________ 10

ii

4

Introduction

As land management agenciescontinue aggressive wildland burning programs, the effects of

increased airborne particulate matterbecome of more concern to landmanagers. Real-time monitoring ofairborne particulates is critical to the landmanager’s ability to protect human health,public safety, and visibility. Landmanagers need reliable methods tomeasure particulate concentrations at thetime of the fire to help make operationaldecisions. Gravimetric techniques requirethat filters collect the particulate matter.The filters are later weighed at specialfacilities routinely used in air qualityapplications. Instruments that inferparticulate concentration by measuringoptical properties of the atmosphere (light

scattering, absorption, or extinction) aremore practical and useful for near real-time measurement.

In March and July of 1998, the MissoulaTechnology and Development Center(MTDC) and the Fire SciencesLaboratory (FSL) evaluated two real-timeoptical instruments that estimateparticulate concentrations by comparingthem with two gravimetric devices. TheMarch laboratory tests were conducted todetermine whether the optical andgravimetric instruments showedsignificant differences when measuringsmoke particles produced from burningwood under controlled conditions in anindoor chamber. Analysis of the data fromthese tests showed considerable

differences between the twomeasurement techniques. Furtherlaboratory tests in July gathered data todevelop a correction curve for the opticalinstruments and to determine if there weresignificant differ-ences in the opticalproperties of smoke based on the type offuel generating it.

All results and conclusions werebased only on laboratory tests.Further testing of the instrumentsin a field environment will beconducted to validate the labora-tory results.

This evaluation was funded by the USDAForest Service’s Watershed, Soil, and AirProgram.

1

5

Instruments

2

Several Forest Service Regions have already purchased optical instruments to measure airborne

particulate levels. Two types of theseinstruments are the MIE DataRAMportable real-time aerosol monitoringsystem and the Radiance ResearchM903 nephelometer. Both instrumentsmeasure the extinction coefficient of lightdue to light backscattering (Bscat).Aerosol levels are commonly reported inmass concentration, usually expressed inmicrograms per cubic meter (ug/m3). Themass concentration is estimated fromBscat by applying a conversion factor.

The DataRAM (Figure 1) is a compact,self-contained instrument that internallyestimates mass concentration from themeasured Bscat (particulatemeasurement ranges from 0.1 ug/m3 to400,000 ug/m3). It continuously displaysthe current and time-weighted averagemass concentration while logging up to10,000 data points. Optionally, theDataRAM can be configured with either aPM-2.5 or PM-10 impactor head toprevent particles larger than 2.5 or 10microns, respectively, from entering the

optical chamber. In these tests, the PM2.5 impactor head was used. For customcalibrations, particulate can be collectedon a filter located in the base of theinstrument. An in-line heater may also beinstalled for monitoring in conditions ofhigh humidity (typically above 70%relative humidity) or fog. The tubularheater is designed to raise thetemperature of the sampled air stream toevaporate liquid water from airborneparticles or to eliminate fog droplets.

The M903 Radiance Research portablenephelometer (Figure 2) is a lightweight,low-power instrument designed forportable operation as well as generalenvironmental monitoring. It measuresand displaysBscat, not massconcentration likethe DataRam.Massconcentrationsmust be calculatedby hand from theBscat readings.The instrumenthas a particulate

measurement range from approximately 1to 1000 ug/m3 when converted fromBscat. The instrument has an internal datalogger that will store scattering coefficientaverages and the operating parametersthat are used to estimate the Bscat. Thestored data can be retrieved using apersonal computer through an RS232port. Different averaging times and logintervals may be set. The instrument canstore approximately 2 weeks of 5-minuteaverages.

We investigated a third instrument, theDustLite, manufactured by Rupprecht andPatachnick. The DustLite estimates massconcentration much like the MIEDataRAM, but the purchase price is a

Figure 1–The MIE DataRam. Figure 2–The Radiance Research nephelometer.

63

fraction of the DataRAM’s price. However,after an initial evaluation, we concludedthat this instrument would not besatisfactory for Forest Service field use.

Gravimetric filter samples of 2.5 um andsmaller were collected concurrently withthe optical instruments to determine actualparticulate levels. For the March tests,two different instruments were used to

gather the gravimetric samples. One wasan Environmental Protection Agency(EPA) Reference Method PM 2.5 samplermanufactured by Rupprecht & PatachnickCo., Inc. (R&P). IML Air Science inSheridan, WY, weighed the 47-mmdiameter Teflon filters used by thisinstrument. The second instrument was aPM 2.5 sampler developed by the FireSciences Laboratory (FSL) for conducting

airborne smoke studies. This instrument’s37-mm diameter filters were weighed in aspecial environmentally controlled facilityat the Fire Sciences Laboratory.Concentrations of particulate inmicrograms per cubic meter werecalculated from flow and gravimetric data.For the July series of tests, only theRMFSL filter system was available.

7

Methods

Figure 347%

4 11/16 x 3 1/810 x 6 3/4

Print to Outside Edge of BordersDO NOT Print Borders

Figure 452%

3 1/2 x 5 3/86 3/4 x 8 1/2

Print to Outside Edge of BordersDO NOT Print Borders

Figure 3–Instrument layout on the smoke platform in the burn chamber.

Figure 4–An example of the firebeds of pine needles used to producesmoke in the laboratory burn chamber.

4

The laboratory study wasconducted in the Fire SciencesLaboratory’s large (131,000-ft3)

combustion chamber. All instrumentswere calibrated and operated on asmoke-sampling platform 55 feet abovethe chamber floor (Figure 3). Theexperiments were performed at ambientconditions inside the closed chamber(70 to 90° F and 30 to 50% relativehumidity). Small beds of flaming andsmoldering ponderosa pine needles onthe chamber floor (Figure 4) generatedsmoke for the majority of tests. Smokefor the remainder of the tests wasgenerated by burning a small amount ofduff. The instruments were close to oneanother so they would be samplingapproximately the same air. Fans helpedmix the smoke in the chamber.

A total of 66 tests were completed inMarch and July. The duration varied foreach test. The higher concentration testswere shorter to prevent the gravimetricfilters from becoming clogged. The lowerconcentration tests were longer to allowenough particulate to accumulate on thefilters for accurate measurements. Theaverage duration of the tests was about1 hour.

8

Results and Discussion

Figure 5–Estimated mass concentration readings of DataRAMs and converted mass concentration readings from the Radiance ResearchNephelometer for Test No. 7 in March 1998.

Particulate Monitor Test - All InstrumentsTest No. 7

0.00

100.00

200.00

300.00

400.00

500.00

600.00

700.00

13:4

8:30

13:5

0:00

13:5

1:30

13:5

3:00

13:5

4:30

13:5

6:00

13:5

7:30

13:5

9:00

14:0

0:30

14:0

2:00

14:0

3:30

14:0

5:00

14:0

6:30

14:0

8:00

14:0

9:30

14:1

1:00

14:1

2:30

14:1

4:00

14:1

5:30

14:1

7:00

14:1

8:30

14:2

0:00

14:2

1:30

14:2

3:00

14:2

4:30

14:2

6:00

14:2

7:30

14:2

9:00

14:3

0:30

14:3

2:00

14:3

3:30

14:3

5:00

14:3

6:30

Time

Inst

rum

ent

Co

nce

ntr

atio

n (

ug

/m3 )

Radiance (Fire Lab)

Radiance (MTDC #1)

Radiance (MTDC #2)

Region 5 DataRAM

Region 2 DataRAM

5

The estimated mass concentrationof the DataRAM and the convertedBscat mass concentrations from

the Radiance Research nephelometersgive similar estimates of real-timeparticulate concentrations. Figure 5shows the particulate concentrationestimates of two DataRAMs and threeRadiance Research nephelometersduring Test No. 7 in March 1998. For thistest, the temperature was 68° F and therelative humidity was 25%. The averagevalue for each instrument was within +/-6% of the mean value for all instruments.

The particulate concentrations measuredby the optical instruments weresignificantly higher than those measuredby the gravimetric instruments for allconcentration ranges and all instruments.Figures 6, 7, and 8 show the respectivenephelometer concentrations plottedagainst the gravimetric concentrations.Concentration estimates for the opticalinstruments were made using conversionfactors supplied by the manufacturer.Based on the best-fit lines, the Region 2DataRAM, MTDC DataRAM, and theRadiance Research nephelometershowed PM 2.5 concentrations 2.1, 1.9,

and 1.9 times higher, respectively, thanthose measured by gravimetric devices.The readings were nearly linear with acorrelation coefficient (R2) of 0.97 andhigher.

Test results were compared whendifferent materials (pine needles and duff)were burned. There was no significantdifference in the instrument readingsbetween the tests. Figure 9 shows theMTDC DataRam mass concentrationreadings for both pine needle and duffsmoke plotted against the gravimetricconcentrations.

9

Best Fit Line for Region 2 DataRAMAll Data - March & July 98

y = 2.1605x

R2 = 0.9722

0

200

400

600

800

1000

1200

0 200 400 600 800 1000 1200

Filter Concentration (u g/m3)

R2

Dat

aRA

M C

on

cen

trat

ion

( ug

/m3 ) Region 2 DataRAM

Linear (Region 2 DataRAM)

Figure 6–Best-fit line for Region 2’s DataRam based on the instrument’s estimated mass concentrations and results from the gravimetric sampler.

6

10

Best Fit Line for MTDC DataRAMAll Data - July 98

y = 1.9342x

R2 = 0.9784

0

100

200

300

400

500

600

700

800

900

0 100 200 300 400 500 600 700 800 900

Filter Concentration (u g/m3)

MT

DC

Dat

aRA

M C

on

cen

trat

ion

(ug

/m3 )

MTDC DataRAM

Linear (MTDC DataRAM)

Figure 7–Best-fit line for MTDC’s DataRam based on the instrument’s estimated mass concentrations and results from the gravimetric sampler.

7

11

Best Fit Line for MTDC Radiance Research NephelometerAll Data - March & July 98

y = 1.9014x

R2 = 0.9715

0

100

200

300

400

500

600

700

800

900

1000

0 100 200 300 400 500 600 700 800 900 1000

Filter Concentration (u g/m3)

Rad

ian

ce R

esea

rch

Co

nce

ntr

atio

n (

ug

/m3 )

Radiance Research

Linear (Radiance Research)

Figure 8–Best-fit line for the Radiance Research nephelometer based on the instrument’s converted mass concentrations and results from thegravimetric sampler.

8

12

Figure 9–Comparison of the DataRam’s estimated mass concentration readings for smoke generated from pine needles and duff compared to resultsfrom the gravimetric sampler.

Comparison of DataRam Readings for Smoke Generated from Pine Needles and Duff

0

100

200

300

400

500

600

700

800

900

1000

0 100 200 300 400 500 600 700 800 900 1000

Gravimetric Concentration (u g/m3)

Inst

rum

ent

Co

nce

ntr

atio

n (

ug

/m3 )

MTDC DataRam (Pine Needles)

MTDC DataRam (Duff)

9

13

Conclusions

10

The MIE DataRAM and RadianceResearch nephelometer bothoverestimated mass

concentration of biomass smoke.Caution is advised when using opticalinstruments to determine massconcentrations of airborne smokeparticulate. More accurate estimates ofparticulate concentration of biomasssmoke can be achieved with the instru-ments tested if a correction factor of 0.53(1/1.9) is applied to the optical reading

(DataRAM concentration * 0.53). Sincethe correction factor is linear, the DataRAMcan be programmed by the user toautomatically output the corrected value.For the Radiance Research instrument,multiply the Bscat conversion factor sug-gested by the vendor by 0.53 when con-verting from Bscat to mass concentration.

For this laboratory study, burning differentbiomass materials did not significantlyaffect the concentration estimates.

The effects of other common atmos-pheric scattering aerosols, such assulfates, nitrates, and soil material werenot evaluated. Our suggested correctionfactor may not be appropriate when theambient aerosol budget is dominated bythese nonbiomass aerosols or evenwhen it includes some sizeable fractionof them.

When the ambient relativehumidity is higher than 70%,optical instruments will have

even greater bias estimating particulatemass concentrations than during dryatmospheric conditions. A high priorityshould be given to further investigatingnephelometer operation in smoky, high-humidity environments. From this work,

Additional Tests

appropriate correction curves and/ormore effective dryers may be developed.

A series of field tests should beconducted using optical instruments andgravimetric instruments. Setting up theinstruments adjacent to a relatively largeprescribed burn would provide a realisticcomparison.

Further tests should be conducted withdifferent fuel types (such as greenbranches and needles, or deciduoustrees) and with different pollutant types(particularly sulfates, nitrates, and soilmaterial) to further identify instrumentcharacteristics.

i

About the AuthorsRonald Babbitt has bachelor’s degrees in forestry and electrical engineering. He joined the Forest Service in 1975 and has worked in areas of forestutilization, forest engineering, and fire research. He now works with the Fire Chemistry Research Project at the Fire Sciences Laboratory in Missoula,MT.

Rich Fisher has bachelor’s degrees in life science (U.S. Air Force Academy) and in meteorology (North Carolina State University). He has master’sdegrees in earth science (Colorado State University) and in international and strategic studies (Naval War College). He is a certified consultingmeteorologist and is Reserve Assistant to the Director of Weather for the U.S. Air Force in the Pentagon.

Harold Thistle received a Ph.D. in plant science specializing in forest meteorology from the University of Connecticut in 1988. He iscertified by the American Meteorological Society as a Certified Consulting Meteorologist (CCM), and worked as a consultant in privateindustry before joining MTDC in 1992. He served as the Center’s Program Leader for Forest Health Protection until 1998, developingmodeling techniques that accurately describe transport of pesticides in the atmospheric surface layer and evaluating meteorologicalinstrument systems for environmental monitoring. He now works with the Forest Health Technology Enterprise Team in Morgantown, WV.

Andy Trent is a Project Engineer at MTDC. He received his bachelor’s degree in mechanical engineering from Montana State University in 1989.He came to MTDC in 1996 and works on projects for the Nursery and Reforestation, Forest Health Protection, and Watershed, Soil, and AirPrograms.

Andria Holland-Sears received her bachelor’s degree in watershed science from Utah State University in 1981. Before moving to Coloradoin 1994, she worked extensively as an Air Resource Specialist for the Forest Service in the Pacific Southwest Region (5) out of Lake Tahoe.Now she works as a hydrologist on the White River National Forest. She also provides technical support related to smoke management andair quality to the Forest Service Regional Office in Denver.

Additional single copies of this document may be ordered from:

USDA Forest ServiceMissoula Technology and Development Center5785 Hwy 10 WMissoula, MT 59808Phone: (406) 329-3900Fax: (406) 329-3719E-mail: wo_mtdc_pubs@fs.fed.us

For further technical information, please contact Harold Thistle or Andy Trent at the address above.

Harold ThistlePhone: (304) 285-1574E-mail: hthistle@fs.fed.us

Andy TrentPhone: (406) 329-3912Fax: (406) 329-3719E-mail: atrent@fs.fed.us