A Comparison Study of Measured Values for Ambient Suspended Particulate Matter Using β Attenuation Method and Gravimetric High-Volume Method —from Intermittent Monitoring Data at Two Sites in Tokyo for a Decade—

Ikuo WATANABE; Osamu ENDO; Sumio GOTO; Kiyoshi TANABE; Tsuguo MIZOGUCHI; Hidetsuru MATSUSHITA

doi:10.5985/jec.10.557

A Comparison Study of Measured Values for Ambient Suspended Particulate Matter Using β Attenuation Method and Gravimetric High-Volume Method

—from Intermittent Monitoring Data at Two Sites in Tokyo for a Decade—

Ikuo WATANABE, Osamu ENDO, Sumio GOTO, Kiyoshi TANABE, Tsuguo MIZOGUCHI, Hidetsuru MATSUSHITA

Author information

Keywords: suspended particulate matter, comparison study, air monitoring, β attenuation method, gravimetric high-volume method

JOURNAL FREE ACCESS

2000 Volume 10 Issue 3 Pages 557-572

DOI https://doi.org/10.5985/jec.10.557

Details

Published: September 29, 2000 Received: May 24, 2000 Available on J-STAGE: May 31, 2010 Accepted: May 24, 2000 Advance online publication: - Revised: -
Correction information
Date of correction: May 31, 2010 Reason for correction: - Correction: ABSTRACT Details: Wrong : Suspended particulate matter (SPM) in the ambient air had been measured with using the two methods, i.e., a β attenuation method (β-ray absorption method; particles larger than 10μm were removed by a cyclon type separator) and a gravimetric high-volume (HV) method (the exposed filters were kept in 50% relative humidity for more than one day before weighing), at two monitoring stations (Ohta and Minato) in Tokyo for a decade (1984-1993) . The measurements were run for 24-hours on randomly selected non-consecutive two days per month.
1) High correlation between the two data sets was observed on the SPM concentrations in each station, [Ohta Sta.; r (correlation coef.) =0.690 (0.529, 0.879) ], [Minato Sta.; r = 0.857 (0.807, 0.907) ], where; the numbers in the parentheses mean the correlation coef. for the periods before and after March in 1991 (when the instruments for a β method were changed) . 2) Linear relationships between the SPM concentrations by an HV and a β method after April in 1991 were;
Ohta Sta.; HV = [1.2-1.4-1.7] ⋅β+ [10-27-34] ………………… (1)
Minato Sta.; HV = [0.9-1.1-1.6] ⋅β+ [3-26-36] …………………… (2)
Where; HV = [α₁-α-α₂] ⋅β+ [γ₁-γ-γ₂],
α&γ: slope & intercept of the line by principal component analysis.
α₁-α₂ & γ₁-γ₂: 95% confidence interval for slope & intercept estimated by a Bootstrap method.
3) The variation patterns of SPM concentration at the two sites were very similar in spite of 8 km distance. 4) SPM concentration was increased under the weather conditions of higher relative humidity and/or weak wind. 5) The difference of values between HV and β was decreased with higher relative humidity (>80%) and/or lower HV value.
In addition, the SPM concentration for HV induced from the relationship showing in eq. (1) and (2) were compared with the HV values in 12 mega-cities in the world.
Right : Suspended particulate matter (SPM) in the ambient air had been measured with using the two methods, i.e., a β attenuation method (β-ray absorption method; particles larger than 10μm were removed by a cyclon type separator) and a gravimetric high-volume (HV) method (the exposed filters were kept in 50% relative humidity for more than one day before weighing), at two monitoring stations (Ohta and Minato) in Tokyo for a decade (1984-1993) . The measurements were run for 24-hours on randomly selected non-consecutive two days per month.
1) High correlation between the two data sets was observed on the SPM concentrations in each station, [Ohta Sta.; r (correlation coef.) =0.690 (0.529, 0.879) ], [Minato Sta.; r = 0.857 (0.807, 0.907) ], where; the numbers in the parentheses mean the correlation coef. for the periods before and after March in 1991 (when the instruments for a β method were changed) . 2) Linear relationships between the SPM concentrations by an HV and a β method after April in 1991 were;
Ohta Sta.; HV = [1.2-1.4-1.7] ·β+ [10-27-34] ………………… (1)
Minato Sta.; HV = [0.9-1.1-1.6] ·β+ [3-26-36] …………………… (2)
Where; HV = [α₁-α-α₂] ·β+ [γ₁-γ-γ₂],
α&γ: slope & intercept of the line by principal component analysis.
α₁-α₂ & γ₁-γ₂: 95% confidence interval for slope & intercept estimated by a Bootstrap method.
3) The variation patterns of SPM concentration at the two sites were very similar in spite of 8 km distance. 4) SPM concentration was increased under the weather conditions of higher relative humidity and/or weak wind. 5) The difference of values between HV and β was decreased with higher relative humidity (>80%) and/or lower HV value.
In addition, the SPM concentration for HV induced from the relationship showing in eq. (1) and (2) were compared with the HV values in 12 mega-cities in the world.

Abstract

Suspended particulate matter (SPM) in the ambient air had been measured with using the two methods, i.e., a β attenuation method (β-ray absorption method; particles larger than 10μm were removed by a cyclon type separator) and a gravimetric high-volume (HV) method (the exposed filters were kept in 50% relative humidity for more than one day before weighing), at two monitoring stations (Ohta and Minato) in Tokyo for a decade (1984-1993) . The measurements were run for 24-hours on randomly selected non-consecutive two days per month.
1) High correlation between the two data sets was observed on the SPM concentrations in each station, [Ohta Sta.; r (correlation coef.) =0.690 (0.529, 0.879) ], [Minato Sta.; r = 0.857 (0.807, 0.907) ], where; the numbers in the parentheses mean the correlation coef. for the periods before and after March in 1991 (when the instruments for a β method were changed) . 2) Linear relationships between the SPM concentrations by an HV and a β method after April in 1991 were;
Ohta Sta.; HV = [1.2-1.4-1.7] ·β+ [10-27-34] ………………… (1)
Minato Sta.; HV = [0.9-1.1-1.6] ·β+ [3-26-36] …………………… (2)
Where; HV = [α₁-α-α₂] ·β+ [γ₁-γ-γ₂],
α&γ: slope & intercept of the line by principal component analysis.
α₁-α₂ & γ₁-γ₂: 95% confidence interval for slope & intercept estimated by a Bootstrap method.
3) The variation patterns of SPM concentration at the two sites were very similar in spite of 8 km distance. 4) SPM concentration was increased under the weather conditions of higher relative humidity and/or weak wind. 5) The difference of values between HV and β was decreased with higher relative humidity (>80%) and/or lower HV value.
In addition, the SPM concentration for HV induced from the relationship showing in eq. (1) and (2) were compared with the HV values in 12 mega-cities in the world.

Corresponding author

Correction information

Register with J-STAGE for free!