FY 1997

Performance characteristics of a high-sensitivity, three-wavelength, total scatter/backscatter nephelometer

Anderson, T.L., D.S. Covert, S.F. Marshall, M.L. Laucks, R.J. Charlson, A.P. Waggoner, J.A. Ogren, R. Caldow, R.L. Holm, F.R. Quant, G.J. Sem, A. Wiedensohler, N.A. Ahlquist, and T.S. Bates

J. Atmos. Ocean. Technol., 13(5), 967–986, doi: 10.1175/1520-0426(1996)013<0967:PCOAHS>2 (1996)

As designed in the 1940s by Beuttell and Brewer, the integrating nephelometer offers a direct method of measuring light scattering by airborne particles without assumptions about particle composition, shape, or physical state. A large number of such instruments have been deployed; however only a limited number of validation experiments have been attempted. This paper reports a set of closure experiments in which a gas-calibrated nephelometer is used to measure the scattering coefficient of laboratory-generated particles of known size and refractive index.

Specifically, it evaluates the performance of a high-sensitivity, three-wavelength, total scatter/backscatter integrating nephelometer (TSI, Inc., model 3563). Sources of uncertainty associated with the gas-calibration procedure, with photo-counting statistics, and with nonidealities in wavelength and angular sensitivity are investigated. Tests with particle-free gases indicate that noise levels are well predicted by photon-counting statistics and that the nephelometer response is linear over a wide range of scattering coefficients. Tests with particles show average discrepancies between measured and predicted scattering of 4%–7%. Error analysis indicates that these discrepancies are within experimental uncertainty, which was dominated by particle generation uncertainty. The simulation of nephelometer response, which is validated by these tests, is used to show that errors arising from nephelometer nonidealities are less than 10% for accumulation-mode or smaller particles (i.e., size distributions for which the volume mean diameter is 0.4 µm or less) and that significant differences exist between the total scatter and backscatter uncertainties. Based on these finding, appropriate applications of the model 3563 nephelometer are discussed.