In-flight characterization of a compact airborne quantum cascade laser absorption spectrometer

Abstract

Abstract. Here, we report on the development of a new quantum cascade laser infrared absorption spectroscopy (QLAS) instrument, the Airborne Tropospheric Tracer In-situ Laser Absorption spectrometer (ATTILA), for atmospheric trace-gas measurements on board of the German High-Altitude Long-range Observatory (HALO) aircraft. Its small and light design makes it suitable for airborne measurements up to approximately 150 hPa of ambient pressure (13–14 km). The dual laser instrument can measure several trace gases simultaneously in two 36.4 m path astigmatic Herriott cells with a data acquisition frequency of 1 Hz. We describe the measurement method and the data acquisition of ATTILA and its in-flight performance by focusing on potential sources of influences on the signal, which we investigated with a dedicated test flight during which the instrument sampled from a constant source. We show that linear critical influences associated with challenging movement patterns can be corrected afterwards, while nonlinear limitations can be minimized by appropriate calibration frequencies and integrated time intervals. During the recent aircraft campaign CAFE Brazil (Chemistry of the Atmosphere Field Experiment in Brazil) from December 2022 to January 2023, carbon monoxide (CO) measurements from ATTILA show a good agreement of a R2 of 0.89 with simultaneous CO measurements from an established IR spectrometer for airborne measurements, the TRacer In Situ TDLAS (tunable diode laser absorption spectroscopy) for Atmospheric Research (TRISTAR), at a 10 s time resolution. First dynamical characteristics and tracer distributions of CO and methane (CH4) over the Amazon rainforest can be identified with ATTILA measurements with a total measurement uncertainty of 10.1 % and 17.5 % for calibration gas mixing ratios of 153 and 1990 ppbv and a data accuracy of 0.3 % and 5.5 % for a data acquisition frequency of 1 Hz for CO and CH4, respectively.