Least Squares Fit Of Line Profiles In Transmittance And Absorbance Spectra With Detector Or Source Noise

Sasada, Hiroyuki

Permalink

https://hdl.handle.net/2142/116567 Copy

Description

Title

Least Squares Fit Of Line Profiles In Transmittance And Absorbance Spectra With Detector Or Source Noise

Author(s)

Sasada, Hiroyuki

Issue Date

2022-06-24

Keyword(s)

Lineshapes, collisional effects

Abstract

When an observed profile of a spectral line is analyzed using an equally weighted least squares method, the noise property of the spectrum determines whether transmittance or absorbance spectrum is appropriate for the analysis. To verify this, we simulate transmittance spectra (TS) of Lorentz profiles with three simulation parameters of absorption strength, center frequency, and width and add either detector noise (DN) or source noise (SN) to the simulated TS. The TSs with DN or SN and absorbance spectra (AS), negative logarithms of them, are fitted to the Lorentz profile using least squares methods. Equally weighted fits of TS with DN and AS with SN, as statistic mathematics predicts, reproduce the noise magnitude and the parameters well and give the expected uncertainties close to the standard deviations of a thousand simulated spectra regardless of the absorption intensity and the noise magnitude. In contrast, equally weighted fits of TS with SN and AS with DN reproduce the simulation parameters but not the noise magnitude and do not predict the uncertainties of the parameters. Properly weighted fits of TS with SN and AS with DN reproduce the noise magnitude and give the expected uncertainties like those given from equally weighted fits of AS with SN and TS with DN but do not always reproduce the absorption strength and width.

Publisher

International Symposium on Molecular Spectroscopy

Type of Resource

text

Language

eng

Handle URL

https://hdl.handle.net/2142/116567&&

DOI

https://doi.org/10.15278/isms.2022.FE03

Copyright and License Information

Copyright 2022 held by the authors

Owning Collections