JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 105, NO. D15, PAGES 19,811-19,822, AUGUST 16, 2000 Comparisonof in situ NzO and CH4 measurements in the upper troposphere and lower stratosphere during STRAT and POLARIS D.F. Hurst, l,2 G.S. Dutton,l,2 P.A. Romashkin, l,2 J.W. Elkins,1R.L. Herman,3 E.J. Moyer, 4D.C. Scott,3 R.D. May,3 C.R. Webster,3 J.Grecu, 5 M. Loewenstein, 5 andJ.R. Podolske • Abstract. Nitrous oxide(N20) and methane (CH4) weremeasured in theupper troposphere and lower stratosphere by multipleinstruments aboard theNASA ER-2 aircraft during the 1995-1996 Stratospheric Tracers of Atmospheric Transport (STRAT) and 1997 Photochemistry of Ozone Loss in theArctic Regionin Summer (POLARIS) campaigns. Differences between coincidental, in situmeasurements are examined to evaluate the agreement andvariabilityin the agreement be- tween these instruments duringeach flight. Mean N20 measurement differences for each flight weremuch smaller thanlimits calculated from quoted values of N20 measurement accuracy and for all but two flightswere <8.7 ppb(3.5%). Mean CH4 measurement differences for flightswere similarly muchsmaller thancalculated limits andfor all but three flightswere<65 ppb(4.4%). Typicalagreement between instruments during flightsaveraged 6.2 ppb(2.5%) for N20 and43 ppb (2.9%) for CH4. In contrast, for about half of the flights,the variability of N20 andCH4 measurement differences exceeded limits calculated from quoted values of measurement preci- sion. The typical measurement difference variability (1•) during a flight averaged _+8.0 ppb (3.2%) for N20 and_+43 ppb (2.9%) for CH4. For some flights,largedifferences or variations in differences areattributable to the poor measurement accuracy or precision of oneinstrument. It is demonstrated that smalloffsets between the computer clocks of these instruments canresultin significant differences between their "coincidental" N20 andCH4 data, especially whenthere is high spatial variabilityin tracer abundance alonga flight track. 1. Introduction Methane(CH4) and nitrousoxide (N20) are tracegases emitted to the atmosphere by natural and anthropogenic surface sources. Despitesignificant removalof CH4 in the troposphere both gases are sufficientlylong-lived to reach the stratosphere where their lifetimesare of the order of 100 years [Volk et al., 1997; World Meteorological Organization (WMO), 1999]. Bothgases have high global warmingpotentials [WMO, 1995] and are accumulating in theatmosphere [Dlugokencky et al., 1994; Prinn et al., 1990]. CH4 oxidation is an important sourceof stratospheric water vapor [Ehhalt and Tonnissen, 1980], and N20 is oxidized in the strato- sphereto produceNO that destroys ozone in a catalytic cycle [Crutzen,1970;McElroy and McConnell, 1971]. In situmeasurements of N20 andCH4 onboard the NASA ER-2 aircraftprovide high-resolution spatial distributions of these long- lived tracers in the upper troposphere and lower stratosphere. Measurements madeduringthe 1995-1996 Stratospheric Tracers of Atmospheric Transport (STRAT) and the 1997 Photochemistry of IClimate Monitoring and Diagnostics Laboratory, National Oceanic and Atmospheric Administration, Boulder, Colorado. 2Cooperative Institute for Research in Environmental Sciences, Univer- sityof Colorado, Boulder. 'Jet Propulsion Laboratory, Pasadena, California. 4California Institute of Technology, Pasadena. •'NASA Ames Research Center, Moffett Field, California. Copyright 2000 by theAmerican Geophysical Union. Paper number 2000JD900218. 0148-0227/00/2000JD900218 $09.00 Ozone Loss in the Arctic Region in Summer (POLARIS) cam- paigns span wide latitude ranges of the Northern Hemisphere during each season. Mean meridional distributions of stratospheric N20 derived from these and other data [Strahan et al., 1999] can be usedto assess and refine transport processes in two-dimensional (2-D) and3-D atmospheric models [Mahlmanet al., 1986]. Meas- urements of CH4 (and H20) from the ER-2 havebeenused to ex- amine the total hydrogen budget of the lower stratosphere [e.g., Dessler et al., 1994; Hurst et al., 1999]. In this work we examine the meanand variabilityof differences between coincidental measurements of N20 and CH4 during each flight. These dataare widely available to the atmospheric science community for studies of stratospheric dynamics and chemistry. The goals of this paper are to (1) investigate whether measurement differences and their variability are within limits calculated from quoted values of the accuracy andprecision of each instrument, (2) identifyflights where the meandifference or variability in differ- ences between instruments is substandard relative to the majority of flights,and (3) to calculate values typicalof the agreement and variability in agreement between instruments during these two missions.It is hoped that investigators employing these N20 and CH 4 data sets will avoidusing the substandard data identified in this work and will incorporate the typicalagreement values in their error analyses. 2. Measurements Three instruments providedthe N20 and CH4 data evaluated in this paper: The NASA Ames Research Center Airborne Tunable LaserAbsorption Spectrometer (ARC/ATLAS), the Jet Propulsion Laboratory Aircraft Laser Infrared Absorption Spectrometer 19,811