Ambient Fine Aerosol Concentrations in Multiple Metrics in Taconite Mining Operations

Tran Huynh, Gurumurthy Ramachandran, Harrison Quick, Jooyeon Hwang, Peter C Raynor, Bruce H Alexander, Jeffrey H Mandel

Research output: Contribution to journalArticle

Abstract

Studies in environmental epidemiology and of occupational cohorts have implicated the effects of fine particulates with increased risk of cardiovascular diseases. Motivated by this evidence, we conducted an ambient air monitoring campaign to characterize fine aerosol concentrations around various taconite ore processes in six taconite mines in northeastern Minnesota. The ore processes were first categorized into 16 broad work areas/buildings. We then took air samples at 91 fixed locations using an array of direct-reading instruments to obtain measurements of mass (PM 2.5 or particles with aerodynamic diameter <2.5 μm, and respirable particulate matter or RPM), alveolar-deposited surface area (ADSA), and particle number (PN) concentrations. At each location, a respirable gravimetric pump (which was used for calibration purposes) and the instruments measured the ambient dust level for 4 h producing ∼240 1-min averaging real-time measurements. To analyze these data, we fit a Bayesian hierarchical model with an autoregressive order 1 correlation structure to estimate pooled concentrations for the 16 work areas/buildings while accounting for temporal correlation. PM 2.5 and RPM average ambient concentrations were highly correlated to each other (Pearson's correlation = 0.98), followed by ADSA and PN correlation (R = 0.77). Office and control room areas were found to have the lowest concentrations in all four metrics when compared to other groups. Distinguishing between concentration levels among the remaining groups was more difficult due to the high uncertainty associated with the geometric mean estimates. The geometric standard deviation within location (GSDWL) generally ranged from 1 to 3 for all exposure metrics, except for a few locations that may have had changes in the work activities that generated the observed peaks and variability during the sampling duration. The geometric standard deviation between locations estimates were generally higher than GSDWL, which may indicate larger variability in the processes/ activities between locations within each broad work area/building. Future work may look into whether it is feasible to use area measurements for epidemiological investigation and use personal measurements (if available) to validate such approach.

Original languageEnglish (US)
Pages (from-to)77-90
Number of pages14
JournalAnnals of Work Exposures and Health
Volume63
Issue number1
DOIs
StatePublished - Jan 7 2019

Fingerprint

Aerosols
Air
Particulate Matter
Dust
Calibration
Uncertainty
Reading
Epidemiology
Cardiovascular Diseases
taconite

Keywords

  • Direct-reading instruments
  • Exposure metrics
  • Fine particulates
  • Taconite mines

PubMed: MeSH publication types

  • Journal Article

Cite this

Ambient Fine Aerosol Concentrations in Multiple Metrics in Taconite Mining Operations. / Huynh, Tran; Ramachandran, Gurumurthy; Quick, Harrison; Hwang, Jooyeon; Raynor, Peter C; Alexander, Bruce H; Mandel, Jeffrey H.

In: Annals of Work Exposures and Health, Vol. 63, No. 1, 07.01.2019, p. 77-90.

Research output: Contribution to journalArticle

Huynh, Tran ; Ramachandran, Gurumurthy ; Quick, Harrison ; Hwang, Jooyeon ; Raynor, Peter C ; Alexander, Bruce H ; Mandel, Jeffrey H. / Ambient Fine Aerosol Concentrations in Multiple Metrics in Taconite Mining Operations. In: Annals of Work Exposures and Health. 2019 ; Vol. 63, No. 1. pp. 77-90.
@article{4aae412a7fe4490395f7bfbaa2e89a9b,
title = "Ambient Fine Aerosol Concentrations in Multiple Metrics in Taconite Mining Operations",
abstract = "Studies in environmental epidemiology and of occupational cohorts have implicated the effects of fine particulates with increased risk of cardiovascular diseases. Motivated by this evidence, we conducted an ambient air monitoring campaign to characterize fine aerosol concentrations around various taconite ore processes in six taconite mines in northeastern Minnesota. The ore processes were first categorized into 16 broad work areas/buildings. We then took air samples at 91 fixed locations using an array of direct-reading instruments to obtain measurements of mass (PM 2.5 or particles with aerodynamic diameter <2.5 μm, and respirable particulate matter or RPM), alveolar-deposited surface area (ADSA), and particle number (PN) concentrations. At each location, a respirable gravimetric pump (which was used for calibration purposes) and the instruments measured the ambient dust level for 4 h producing ∼240 1-min averaging real-time measurements. To analyze these data, we fit a Bayesian hierarchical model with an autoregressive order 1 correlation structure to estimate pooled concentrations for the 16 work areas/buildings while accounting for temporal correlation. PM 2.5 and RPM average ambient concentrations were highly correlated to each other (Pearson's correlation = 0.98), followed by ADSA and PN correlation (R = 0.77). Office and control room areas were found to have the lowest concentrations in all four metrics when compared to other groups. Distinguishing between concentration levels among the remaining groups was more difficult due to the high uncertainty associated with the geometric mean estimates. The geometric standard deviation within location (GSDWL) generally ranged from 1 to 3 for all exposure metrics, except for a few locations that may have had changes in the work activities that generated the observed peaks and variability during the sampling duration. The geometric standard deviation between locations estimates were generally higher than GSDWL, which may indicate larger variability in the processes/ activities between locations within each broad work area/building. Future work may look into whether it is feasible to use area measurements for epidemiological investigation and use personal measurements (if available) to validate such approach.",
keywords = "Direct-reading instruments, Exposure metrics, Fine particulates, Taconite mines",
author = "Tran Huynh and Gurumurthy Ramachandran and Harrison Quick and Jooyeon Hwang and Raynor, {Peter C} and Alexander, {Bruce H} and Mandel, {Jeffrey H}",
year = "2019",
month = "1",
day = "7",
doi = "10.1093/annweh/wxy086",
language = "English (US)",
volume = "63",
pages = "77--90",
journal = "Annals of Work Exposures and Health",
issn = "2398-7308",
publisher = "Oxford University Press",
number = "1",

}

TY - JOUR

T1 - Ambient Fine Aerosol Concentrations in Multiple Metrics in Taconite Mining Operations

AU - Huynh, Tran

AU - Ramachandran, Gurumurthy

AU - Quick, Harrison

AU - Hwang, Jooyeon

AU - Raynor, Peter C

AU - Alexander, Bruce H

AU - Mandel, Jeffrey H

PY - 2019/1/7

Y1 - 2019/1/7

N2 - Studies in environmental epidemiology and of occupational cohorts have implicated the effects of fine particulates with increased risk of cardiovascular diseases. Motivated by this evidence, we conducted an ambient air monitoring campaign to characterize fine aerosol concentrations around various taconite ore processes in six taconite mines in northeastern Minnesota. The ore processes were first categorized into 16 broad work areas/buildings. We then took air samples at 91 fixed locations using an array of direct-reading instruments to obtain measurements of mass (PM 2.5 or particles with aerodynamic diameter <2.5 μm, and respirable particulate matter or RPM), alveolar-deposited surface area (ADSA), and particle number (PN) concentrations. At each location, a respirable gravimetric pump (which was used for calibration purposes) and the instruments measured the ambient dust level for 4 h producing ∼240 1-min averaging real-time measurements. To analyze these data, we fit a Bayesian hierarchical model with an autoregressive order 1 correlation structure to estimate pooled concentrations for the 16 work areas/buildings while accounting for temporal correlation. PM 2.5 and RPM average ambient concentrations were highly correlated to each other (Pearson's correlation = 0.98), followed by ADSA and PN correlation (R = 0.77). Office and control room areas were found to have the lowest concentrations in all four metrics when compared to other groups. Distinguishing between concentration levels among the remaining groups was more difficult due to the high uncertainty associated with the geometric mean estimates. The geometric standard deviation within location (GSDWL) generally ranged from 1 to 3 for all exposure metrics, except for a few locations that may have had changes in the work activities that generated the observed peaks and variability during the sampling duration. The geometric standard deviation between locations estimates were generally higher than GSDWL, which may indicate larger variability in the processes/ activities between locations within each broad work area/building. Future work may look into whether it is feasible to use area measurements for epidemiological investigation and use personal measurements (if available) to validate such approach.

AB - Studies in environmental epidemiology and of occupational cohorts have implicated the effects of fine particulates with increased risk of cardiovascular diseases. Motivated by this evidence, we conducted an ambient air monitoring campaign to characterize fine aerosol concentrations around various taconite ore processes in six taconite mines in northeastern Minnesota. The ore processes were first categorized into 16 broad work areas/buildings. We then took air samples at 91 fixed locations using an array of direct-reading instruments to obtain measurements of mass (PM 2.5 or particles with aerodynamic diameter <2.5 μm, and respirable particulate matter or RPM), alveolar-deposited surface area (ADSA), and particle number (PN) concentrations. At each location, a respirable gravimetric pump (which was used for calibration purposes) and the instruments measured the ambient dust level for 4 h producing ∼240 1-min averaging real-time measurements. To analyze these data, we fit a Bayesian hierarchical model with an autoregressive order 1 correlation structure to estimate pooled concentrations for the 16 work areas/buildings while accounting for temporal correlation. PM 2.5 and RPM average ambient concentrations were highly correlated to each other (Pearson's correlation = 0.98), followed by ADSA and PN correlation (R = 0.77). Office and control room areas were found to have the lowest concentrations in all four metrics when compared to other groups. Distinguishing between concentration levels among the remaining groups was more difficult due to the high uncertainty associated with the geometric mean estimates. The geometric standard deviation within location (GSDWL) generally ranged from 1 to 3 for all exposure metrics, except for a few locations that may have had changes in the work activities that generated the observed peaks and variability during the sampling duration. The geometric standard deviation between locations estimates were generally higher than GSDWL, which may indicate larger variability in the processes/ activities between locations within each broad work area/building. Future work may look into whether it is feasible to use area measurements for epidemiological investigation and use personal measurements (if available) to validate such approach.

KW - Direct-reading instruments

KW - Exposure metrics

KW - Fine particulates

KW - Taconite mines

UR - http://www.scopus.com/inward/record.url?scp=85059502525&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059502525&partnerID=8YFLogxK

U2 - 10.1093/annweh/wxy086

DO - 10.1093/annweh/wxy086

M3 - Article

VL - 63

SP - 77

EP - 90

JO - Annals of Work Exposures and Health

JF - Annals of Work Exposures and Health

SN - 2398-7308

IS - 1

ER -