Quantitative analysis of temporal stability and instrument performance during field experiments of an airborne QCLAS via Allan–Werle-plots

Abstract

Allan–Werle-plots are an established tool in infrared absorption spectroscopy to quantify temporal stability, maximum integration time and best achievable precision of a measurement instrument. In field measurements aboard a moving platform, however, long integration times reduce time resolution and smooth atmospheric variability. A high accuracy and time resolution are necessary as well as an appropriate estimate of the measurement uncertainty. In this study, Allan-Werle-plots of calibration gas measurements are studied to analyze the temporal characteristics of a Quantum Cascade Laser Absorption Spectrometer (QCLAS) instrument for airborne operation. Via least-squares fitting the individual noise contributions can be quantified and different dominant regimes can be identified. Through simulation of data according to the characteristics from the Allan-Werle-plot, the effects of selected intervals between in-flight calibrations can be analyzed. An interval of 30\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$30\\,$$\\end{document}min is found sufficient for successful drift correction during ground operation. The linear interpolation of the sensitivity increases the accuracy and lowers the measurement uncertainty from 1.1%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$1.1\\,\\%$$\\end{document} to 0.2%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$0.2\\,\\%$$\\end{document}. Airborne operation yields similar results during segments of stable flight but suffers from additional flicker and sinusoidal contributions. Simulations verify an appropriate interval of 30\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$30\\,$$\\end{document}min in airborne operation. The expected airborne measurement uncertainty is 2.45 ppbv.