Airborne intercomparison of HOx measurements using laser-induced fluorescence and chemical ionization mass spectrometry during ARCTAS

摘要

The hydroxyl (OH) and hydroperoxyl (HO) radicals, collectively calledHO, play central roles in tropospheric chemistry. Accuratemeasurements of OH and HO are critical to examine our understanding ofatmospheric chemistry. Intercomparisons of different techniques fordetecting OH and HO are vital to evaluate their measurementcapabilities. Three instruments that measured OH and/or HO radicalswere deployed on the NASA DC-8 aircraft throughout Arctic Research of theComposition of the Troposphere from Aircraft and Satellites (ARCTAS) in thespring and summer of 2008. One instrument was the Penn State AirborneTropospheric Hydrogen Oxides Sensor (ATHOS) for OH and HO measurementsbased on Laser-Induced Fluorescence (LIF) spectroscopy. A second instrumentwas the NCAR Selected-Ion Chemical Ionization Mass Spectrometer (SI-CIMS)for OH measurement. A third instrument was the NCAR Peroxy Radical ChemicalIonization Mass Spectrometer (PeRCIMS) for HO measurement. Formalintercomparison of LIF and CIMS was conducted for the first time on a sameaircraft platform. The three instruments were calibrated by quantitativephotolysis of water vapor by ultraviolet (UV) light at 184.9 nm with three differentcalibration systems. The absolute accuracies were ±32% (2σ) for the LIF instrument, ±65% (2σ) for the SI-CIMSinstrument, and ±50% (2σ) for the PeRCIMS instrument. Ingeneral, good agreement was obtained between the CIMS and LIF measurementsof both OH and HO measurements. Linear regression of the entire dataset yields [OH] = 0.89 × [OH] + 2.8 × 10 cmwith a correlation coefficient = 0.72 for OH,and [HO] = 0.86 × [HO] + 3.9 partsper trillion by volume (pptv, equivalent to pmol mol) with acorrelation coefficient = 0.72 for HO. In general, thedifference between CIMS and LIF instruments for OH and HO measurementscan be explained by their combined measurement uncertainties. Comparisonwith box model results shows some similarities for both the CIMS and LIFmeasurements. First, the observed-to-modeled HO ratio increasesgreatly for higher NO mixing ratios, indicating that the model may notproperly account for HO sources that correlate with NO. Second, theobserved-to-modeled OH ratio increases with increasing isoprene mixingratios, suggesting either incomplete understanding of isoprene chemistry inthe model or interferences in the measurements in environments wherebiogenic emissions dominate ambient volatile organic compounds.