NILU logo 
Data Submission Manual
SoftwareClosure tests for aerosol data    October 17, 2019
Closure studies for particle light scattering

Checking closure between calculated particle light scattering coefficient and measured particle light scattering coefficient

The aerosol particle light-scattering coefficient is one of the core observation variables in ACTRIS. When considering a light-beam incident into an aerosol volume, it describes the amount of light scattered by the aerosol particles away from its original direction. The particle light-scattering coefficient is essential for describing the direct climate effect of atmospheric aerosol, i.e. its climate effect by scattering incoming solar radiation back into space before it can warm the Earth’s surface. At ACTRIS surface in situ stations, the particle light-scattering coefficient is measured directly by integrating nephelometers (Heintzenberg & Charlson, 1996). The particle number size distribution (PNSD) is another ACTRIS core variable measured at the same stations. It quantifies the number of aerosol particles as a function of particle size. The PNSD is measured by Mobility Particle Size Spectrometers (MPSS) (Wiedensohler et al., 2012) in the size range of particle diameter 0.01 µm < Dp < 0.8 µm. In the majority of cases, this size range covers most of the particles contributing to the particle light-scattering coefficient. 

Meeting the requirements for closure exercises described above, ACTRIS has established operating procedures and traceable, independent calibration chains for both the nephelometer and MPSS systems (Wiedensohler et al., 2012). This includes thresholds for the expected accuracies, which for MPSS measurements are 3.5% deviation from the standard for particle size, and 10% for particle concentration. For nephelometers, the corresponding number is 10% deviation from the standard (Anderson et al., 1996). 
For spherical and well-mixed particles of known chemical composition, the particle scattering coefficient can be calculated from theory for a given PNSD (Bohren & Huffman, 1983). The systematic uncertainty in the calculated scattering coefficient caused by assuming a particle chemical composition, i.e. refractive index real part, is typically around 13% in the ambient atmospheric boundary later (Fiebig et al., 2002), and much smaller for dried aerosol as dealt with at ACTRIS near-surface stations. Thus, deviations between scattering coefficients measured directly and calculated from MPSS will predominantly be caused by uncertainties in the measured PNSD, which includes the size range of particles larger than the upper size limit of the MPSS. However, this size range becomes relevant only for certain aerosol types with high particle concentrations in this size range, e.g. marine and dust aerosols. As a result, closure between the particle scattering coefficients measured by integrating nephelometer and calculated from the PNSD measured by the MPSS a very meaningful exercise to verify the accuracy of both measurements.

The calculation of particle light scattering coefficient

To be suitable for the aerosol particle light scattering closure test, a station needs to operate an integrating nephelometer and a MPSS system, both following the ACTRIS operating procedures and calibration routines. For obtaining regular closure reports, both instruments need to participate in the ACTRIS near-real-time (NRT) data submission and dissemination infrastructure. 
Despite some mathematical sophistication in the detail, calculating the particle light scattering coefficient from the PNSD for comparison with the nephelometer measured scattering coefficient is straight forward in principle. Sampling points are defined across the range of the measured PNSD using homogeneous logarithmic spacing in particle diameter. At the sampling points, the particle scattering cross-section is calculated by means of the Mie-scattering code of Bohren & Huffman (1983). This implies the assumption of homogeneously-mixed spherical particles. After multiplying the scattering-cross section with the particle concentration at the PNSD sampling points, the contributions of all particle sizes are summed up to yield the particle scattering coefficient. This amounts to a numerical integration of the product of particle concentration and scattering cross-section over particle size. The density of sampling points is chosen to keep the calculation uncertainty < 1% while maximising calculation speed.
For obtaining a wavelength-dependent particle refractive index to be used in the calculation, the current algorithm assumes the particles to consist of ammonium sulphate. This substance is not only abundant in atmospheric aerosol particles. Its refractive index real part, quantifying the capacity of a substance to scatter, is also close to that of many other substances commonly encountered in these particles (Fiebig et al., 2002). Water uptake by the particles would change the refractive index significantly, but can the neglected since the ACTRIS operating procedure requires drying of the aerosol to relative humidities < 40%. The closure scheme does currently not assume any absorbing substances in the aerosol in order to minimise the number of instruments and dependencies for the closure scheme. However, absorbing substances in the aerosol such as black carbon (BC) or absorbing mineral dust influence particle scattering for certain aerosol types. An extension to include particle absorption in the closure scheme, represented by BC, is therefore planned for the near future. The extended version will assume BC as absorbing component in the aerosol particle size range covered by the MPSS, represent the BC optical properties by “soot” as described by d’Almeida (1991), and constrain the BC content by collocated observations of the particle absorption coefficient.
Executable versions of the closure software will be made available here in the near future.

Current solution for reporting of particle light scattering coefficient

The automatic intercomparison of the measured and calculated particle light scattering coefficient is currently implemented for stations featuring the required observations and participating in the NRT data submission scheme in order to test for closure, and consists of a collection of processes and software. 
The intercomparison of the measured and calculated particle light scattering coefficient is delivered as a report to data originator(s) and data submitter(s) responsible for the measurements at that station. The setup, including the software and database, is hosted by NILU - Norwegian Institute for Air Research. This includes: software for calculating the scattering coefficient from the MPSS particle number size distribution, software for importing the data into the EBAS database as a separate data product (level 3 data), software for extraction of data from EBAS, routines for plotting comparison plots of the closure variables, and automatic report generation, as well as an automatic e-mail service for regular delivery of reports.

Automatically generated reports are produced and delivered for each participating ACTRIS station, either on a monthly basis for Near-Real-Time (NRT) data, or yearly as a part of the quality control procedures performed for all stations. This will be implemented as a part of the annual data QA/QC procedures. The distribution and frequency of the reports are flexible, can be adapted to the needs and requests from the data originators, station operators or ACTRIS calibration centres. All plots and relevant metadata are consolidated into a PDF document, and then sent as an e-mail attachment. The reports are generated based on each respective dataset, and sent to the data originator(s) and/or data submitter(s), and/or instrument principal investigator(s) listed for the specific dataset. 

Examples from the reporting tool:

The entire deliverable, could be found here: WP3_D3 13_M28_v2.pdf


Anderson, T. L., Covert, D. S., Marshall, S. F., Laucks, M. L., Charlson, R. J., Waggoner, A. P., Ogren, J. A., Caldow, R., Holm, R. L., Quant, F. R., Sem, G. J.,Wiedensohler, A., Ahlquist, N. A., and Bates, T. S. (1996). Performance Characteristics of a High-Sensitivity, Three-Wavelength, Total Scatter/Backscatter Nephelometer, J. Atmos. Ocean. Tech. 13:967–986.
Bohren, Craig F.; Huffman, Donald R. (1983): Absorption and scattering of light by small particles, New York: Wiley.
Cragin, Melissa; Heidorn, P. Bryan; Palmer, Carole L.; Smith, Linda C. (2007). "An Educational Program on Data Curation". ALA Science & Technology Section Conference. Retrieved 7 October 2013.
d’Almeida, G., P. Koepke, and E. Shettle, Atmospheric Aerosols: Global Climatology and Radiative Characteristics, A. Deepak, Hampton, Va., 1991.
Heintzenberg, J. and Charlson, R.J., "Design and applications of the integrating nephelometer: A review," J. Atmospheric and Oceanic Technology, 13, 987-1000, 1996.
 Copyright 2009 by DotNetNuke Corporation   Terms Of Use  Privacy Statement