Abstract
Please click here to find the Original Article to which this Letter refers: http://dx.doi.org/10.1080/027868299304435
INTRODUCTION
The Particle/Soot Absorption Photometer (PSAP, Radiance Research, Seattle, WA) is a widely used instrument for measuring light absorption by aerosols. Its operation is based on the measurement of a change in light transmission through a quartz-fiber filter as particles are deposited on the filter. Bond et al. (1999, hereafter called B1999) described the instrument and derived a scheme for correcting the measurements for systematic errors in the method. This comment continues the discussion in B1999 of wavelength adjustment and clarifies a likely error in the application of the filter area correction. Investigators applying the B1999 correction factors to PSAP data may need to multiply the absorption coefficient reported by the PSAP by an additional factor of 0.873, depending on how they calculate the adjusted absorption coefficient.
FILTER AREA CORRECTION
The PSAP reports a value for the absorption coefficient (σ PSAP ) that is calculated from the primary measurements of filter transmittance at times t and t + Δ t (I(t) and I(t + Δ t)), the sample flow rate Q PSAP measured with the internal flowmeter, and the assumed area (17.83 mm2) of the deposit of particles on the filter, A PSAP :
B1999 point out that variations among instruments require corrections for the true flowrate and filter spot area (Q true and A true ), which must be determined experimentally for each instrument. Combining eqn. 1 above with Equations (Equation4–6) of B1999 yields the adjusted absorption coefficient that B1999 used to derive their PSAP correction factors:
The relationship between the value of the adjusted absorption coefficient used by B1999 and the alternative definition (Equation (Equation4)) is
This factor of 0.873, which is the ratio of the filter spot area used internally by the PSAP (A PSAP = 17.83 mm2) and the measured spot area of the manufacturer's reference instrument (A ref = 20.43 mm2), was briefly described by CitationSheridan et al. (2005), but repeated communications with PSAP users in subsequent years reveal that considerable confusion about the applicability of this correction factor remains. PSAP users need to be aware that the parameters of the B1999 PSAP correction scheme (B1999, Equations (Equation1) and (12)) are based on Equation (Equation3)
B1999 reported the numerical values of K 1 and K 2 to be 0.02 ± 0.02 and 1.22 ± 0.20, respectively.
Equations (Equation7) and (Equation8) yield the same result and are equally valid, the only difference being whether the spot area of the PSAP firmware (A PSAP = 17.83 mm2) or the manufacturer's reference PSAP (A ref = 20.43 mm2) is used in the calculation. Users who calculate the absorption coefficient from the measured intensities, as well as users of modern PSAP's that allow adjustment of the filter area A PSAP in the instrument setup screens, should use Equation (Equation7) to calculate the absorption coefficient.
WAVELENGTH ADJUSTMENT
The scattering and extinction measurements used by B1999 to derive the K 1 and K 2 correction factors were made at a wavelength of 550 nm, while the PSAP measurements were at a slightly longer wavelength. As a result, the B1999 correction factors include an implicit adjustment from the operating wavelength of the PSAP to 550 nm, a fact acknowledged in B1999. The operating wavelength of the PSAP was claimed to be 567 nm by B1999, while a later study (CitationVirkkula et al. 2005, hereafter called V2005) measured the operating wavelength with a spectrophotometer and reported a value of 574 nm.
The Radiance Research PSAP is now available in a three-wavelength model, operating at wavelengths of 467, 530, and 660 nm. A slightly modified version of this instrument was used by CitationSchmid et al. (2006), who corrected the results at all three wavelengths using the B1999 scheme. The light source of the three-wavelength Radiance Research PSAP is different from the light sources used in the original one-wavelength instrument and the prototype three-wavelength instrument characterized by V2005. Because of differences in the light source, detectors, and flow path, it is not obvious that either the B1999 or V2005 correction schemes apply to the three-wavelength Radiance Research PSAP. If the B1999 correction scheme is used, however, the implicit wavelength adjustment must be removed prior to applying the corrections at other wavelengths.
Explicitly including the measurement wavelengths in Equation (Equation7), and using the measured PSAP wavelength of 574 nm, yields the adjusted absorption coefficient at 550 nm wavelength,
Observed particle diameters were from 0.1 to 0.5 μ m in diameter in the B1999 experiments. The wavelength dependence of the absorption coefficient of 0.4 μ m diameter nigrosin particles was calculated by B1999 to be λ–0.5, with the exponent reaching –1 for particles much smaller than the wavelength of light. For a wavelength dependence of λ–0.5, the additional correction term in Equation (Equation10) is
If the wavelength dependence is λ–1, the additional correction factor drops to 0.957. B1999 did not report the size distribution of the particles used in their tests, which precludes calculation of the wavelength dependence of absorption of their polydisperse nigrosin test aerosol. It is reasonable to assume, though, that the additional correction factor is 0.97 ± 0.01.
Equation (Equation11) assumes that the wavelength dependence of the absorption measured by the PSAP is the same as the true wavelength dependence of the nigrosin calibration aerosols. This is not strictly true, because of the contribution of scattering to the PSAP absorption (Equation (Equation9)). However, B1999 reported that the single-scattering albedo of their nigrosin calibration aerosols was 0.5; for such strongly absorbing aerosols, the scattering term in Equation (Equation9) is small and the wavelength dependence of the absorption measured by the PSAP is nearly the same as the true wavelength dependence. For a value of the wavelength adjustment factor of 0.97, the B1999 correction for measurements of scattering and absorption at wavelength λ becomes
The wavelength range over which Equation (Equation12) applies is unknown. Absorption coefficients for the modified 3-wavelength PSAP used by V2005, when calculated using the B1999 correction scheme, were greater than the reference absorption by a similar amount (17–26%) at all three wavelengths. The results in Figure 2 of V2005 suggest that the B1999 correction factors are not strongly dependent on wavelength, at least for the wavelengths used in the 3-wavelength PSAP, and a recently-discovered error in V2005 (CitationVirkkula 2010) resolves a long-standing discrepancy between the 3-wavelength and 1-wavelength PSAPs in V2005.
CONCLUSION
Additional experiments are needed to establish the wavelength range where the B1999 values of K1 and K2 in Equation (Equation12) are valid. It is clear, however, that the wavelength adjustment implicit in the B1999 correction scheme is not directly applicable to PSAPs operating at wavelengths other than 574 nm. Perhaps more importantly, PSAP users need to be aware of the potential for confusion introduced by the manufacturer's use of a filter area different from the value measured by CitationBond et al. (1999). While the wavelength correction is a 3% adjustment, improper treatment of the calibration instrument's filter area adjustment leads to a 13% error in the reported absorption coefficient.
Acknowledgments
This article not subject to United States copyright law.
Related Research Data
REFERENCES
- Bond , T. C. , Anderson , T. L. and Campbell , D. 1999 . Calibration and Intercomparison of Filter-Based Measurements of Visible Light Absorption by Aerosols . Aerosol Sci. Technol. , 30 : 582 – 600 .
- Schmid , B. , Ferrare , R. , Flynn , C. , Elleman , R. , Covert , D. , Strawa , A. , Welton , E. , Turner , D. , Jonsson , H. , Redemann , J. , Eilers , J. , Ricci , K. , Hallar , A. G. , Clayton , M. , Michalsky , J. , Smirnov , A. , Holben , B. and Barnard , J. 2006 . How well do state-of-the-art techniques measuring the vertical profile of tropospheric aerosol extinction compare? . J. Geophys. Res. , 111 : D05S07 doi:10.1029/2005JD005837
- Sheridan , P. J. , Arnott , W. P. , Ogren , J. A. , Andrews , E. , Atkinson , D. B. , Covert , D. , Moosmüller , H. , Petzold , A. , Schmid , B. , Strawa , A. W. , Varma , R. and Virkkula , A. 2005 . The Reno Aerosol Optics Study: An Evaluation of Aerosol Absorption Measurement Methods . Aerosol Sci. Technol. , 39 : 1 – 16 .
- Virkkula , A. , Ahlquist , N. C. , Covert , D. S. , Arnott , W. P. , Sheridan , P. J. , Quinn , P. K. and Coffman , D. J. 2005 . Modification, Calibration and a Field Test of an Instrument for Measuring Light Absorption by Particles . Aerosol Sci. Technol. , 39 : 68 – 83 .
- Virkkula , A. 2010 . Correction of the calibration of the 3-wavelength Particle Soot Absorption Photometer (3λPSAP) . Aerosol Sci. Technol. , 44 : 706 – 712 .