J. D. Shetter

Indications of photochemical histories of Pacific air masses from measurements of atmospheric trace species at Point Arena, California

Measurements of light hydrocarbons, ozone, peroxyacetyl nitrate (PAN), HNO3, NO3−, NOx, NOy, and meteorological parameters were made during a 10-day period in April and May 1985 at Point Arena, a coastal inflow site on the Pacific Ocean in northern California. The meteorological measurements indicate that during this study the sampled air was usually from the marine boundary layer with little land influence on the meteorological parameters. In this marine air the mixing ratios of the alkanes, ozone, PAN, and HNO3 showed strong correlations coincident with variations in the origins of calculated air parcel trajectories and with variations in the ratios of the light alkanes. This variation in the ratios is attributed to different degrees of photochemical aging of the alkanes that are generally consistent with the calculated trajectories. This behavior indicates that the alkane levels are determined by transport to the marine area from continental sources, most likely Asian, followed by photochemical removal over the Pacific Ocean. Since the concentrations of PAN and ozone correlate well with the alkane ratios, it is concluded that the observed PAN and ozone were dominated by continental sources and removal processes in the marine areas. This and other marine studies have observed a strong correlation of PAN and ozone, and it is suggested that production over the continents, transport to the marine areas, and parallel removal processes account for much of the observed correlation. From the correlation of these two species with the measured alkane ratios, approximate net lifetimes of PAN and ozone in the marine troposphere of ≤2.5 and ≥19 days, respectively, are derived. The primary conclusion is that the alkanes, ozone, and PAN in these air parcels from the Pacific marine troposphere are dominated by transport from continental sources and removal by photochemical processes. Direct emissions of the alkanes and in situ photochemical production of PAN and ozone from precursors emitted into the marine region from the surface or the stratosphere must play less important roles. Similar indications of continental influence in marine areas have been seen in other studies of ozone, the sulfur cycle, oxidized nitrogen, and hydrocarbons. It is suggested that the ratios of the light alkanes provide photochemical “clocks” that are useful for gauging the importance of continental influence in a particular marine measurement.

Observations of peroxyacetyl nitrate, peroxypropionyl nitrate, methyl nitrate and ozone during the Mauna Loa Observatory photochemistry experiment

Measurements of the title species were made during the Mauna Loa Observatory Photochemistry Experiment (MLOPEX) conducted between May 1 and June 4, 1988, at the Geophysical Monitoring for Climatic Change (GMCC) station at 3.4-km elevation on the Island of Hawaii. Diurnal changes in the organic nitrates primarily resulted from the transition between downslope flow (usually free tropospheric air) and upslope flow (marine boundary layer or a mix of marine boundary layer and free tropospheric air, both influenced by island sources of precursors) characteristic of the site. Longer term trends in the mixing ratios reflected changes in air mass origins from mid-latitudes to more tropical latitudes. The average mixing ratios in free tropospheric samples were peroxyacetyl nitrate (PAN, 17 pptv), peroxypropionyl nitrate (PPN, 0.3 pptv), methyl nitrate (MN, 4 pptv), and O3 (43 ppbv). The organic nitrates (PAN, PPN, MN) represent minor components of the total odd nitrogen budget at the site. In free tropospheric samples, PAN, PPN, and MN constituted average percentages of 7%, <1%, and 2% of total odd nitrogen. In more tropical air masses, MN could constitute as much as 10% of total odd nitrogen. A photochemical model is used to investigate the sensitivity of free tropospheric PAN to local precursor concentrations. The observed mixing ratios of PAN are also contrasted with measurements made at continental surface sites and during aircraft programs.

Reply [to “Comment on ‘Indications of photochemical histories of Pacific air masses from measurements of atmospheric trace species at Point Arena, California’ by D. D. Parrish et al.”]

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 98, NO. D8, PAGES 14,995-14,997, AUGUST Reply D. D. PARRISH, C. J. HAHN, 1 E. J. WILLIAMS, 1 R. B. NORTON, and F. C. FEHSENFELD Aeronomy Laboratory, National Oceanic and Atmospheric Administration, Boulder, Colorado H. B. SINGH Earth System Science Division, NASA Ames Research Center, Moffett Field, California J. D. SHETTER,2 B. W. GANDRUD, and B. A. RIDLEY National Center for Atmospheric Research, Boulder, Colorado Chlorine atom chemistry in the marine boundary layer is a topic that has received considerable recent attention [see Finlayson-Pitts, this issue, and references therein]. In her comment, Finlayson-Pitts has pointed out that the observed slope of the n-butane to ethane versus propane to ethane correlation in Figure 9 of Parrish et al. [1992] is in closer agreement with C1 kinetics than the OH kinetics that was considered there and that this agreement may indicate that C1 atoms make a substantial contribution to the removal Fraserdale, Ontario (latitude 50 o N, longitude 82 o W), a remote boreal site in eastern central Canada; for this continental data set they found a slope of 1.44 with an r2 of 0.96. All of these slopes are in remarkable agreement, and all are significantly smaller than predicted by OH kinetics, which is 2.75 + 0.55 (confidence limit calculated from 2c• confidence limits of rate constants (R. Talukdar and A. R. Ravishankara, private communication, 1993) and a 250 to 298 K temperature range). of the These relatively small slopes are characteristic of the free light alkanes from the tropical and temperate marine troposphere and continental air masses as well as the marine troposphere. However, we believe that much more substantial boundary layer. Yet, the marine boundary layer is the only evidence must be presented for this conjecture before it can be region of the troposphere where the C1 atom sources cited by accepted. Moreover, evidence presented below suggests that C1 Finlayson-Pitts [this issue] could conceivably lead to a substantial contribution to alkane removal. atom chemistry is unlikely to play a significant role, at least in Second, there are correlation plots of alkane concentration the removal of light alkanes from either the continental or the ratios that can give more direct information regarding the marine troposphere in tropical and temperate regions. As discussed semiquantitatively in the appendix of Parrish et relative contributions of C1 atom and OH radical reactions to alkane removal, regardless of the effects of mixing. Table 1 al. [1992] and earlier by McKeen et al. [1990], the mixing of air masses of different ages leads to a systematic deviation from gives kinetic data for the reactions of C2-C4 alkanes with C1 and the slope expected from the kinetics of the removal processes. OH. The values of the relevant rate constants are such that the Recently, S. A. McKeen and S.C. Liu (Hydrocarbon ratios and C1 reactions will change the ratio of the two isomers of butane photochemical history of air masses, submitted to Geophysical but will leave the/-butane to propane ratio relatively constant. In contrast, OH reactions will leave the ratio of the butanes Research Letters, 1993) have quantitatively reproduced the observed slope from a three-dimensional model of the relatively constant but will change the /-butane to propane chemistry and mixing in the troposphere over the western ratio. Thus a log-log plot of the ratio of/-butane to n-butane Pacific region. This model included only OH removal of the versus /-butane to propane should provide two approximately alkanes. If the effects of mixing are not treated adequately, the orthogonal axes that will separate the C1 and OH contributions slopes of plots such as Figures 9 and 10 of Parrish et al. [1992] cannot be utilized to distinguish between the roles of different removal processes. Nevertheless, two arguments lead us to conclude that C1 atom chemistry plays no more than a minor role in the removal of the light alkanes from the tropical or temperate marine boundary layer. First, Figure 10 of Parrish et al. [1992] includes continental and free troposphere data as well as marine data. All of these data are reasonably well fit ( r 2 = 0.89) by a straight line with a slope (1.47) similar to that found in the Point Arena data alone (1.33) and in the North Atlantic (1.66) by Rudolph and Jobhen [1990]. Recently, Jobson et al. [1993] have presented a similar plot for data collected at 1Also at Cooperative Institute for Research in Environmental Sciences, University of Colorado, Boulder. 2Presently at 841 Paragon Drive, Boulder, Colorado. Copyright 1993 by the American Geophysical Union. Paper number 93JD01416. 0148-0227/93/93 JD-01416502.00 to alkane removal. Furthermore, if one of these two ratios does indeed remain constant, then mixing of two air masses of different ages cannot affect that ratio since it would have the same value in both. Of course, this assumes that the source ratios are relatively constant; the data presented below suggest that this is indeed true. Figure 1 compares experimental data with the behavior expected from the reaction kinetics under five scenarios with differing contributions from C1 and OH reactions. (The format chosen for the plot is log-log rather than a linear plot of the logs of the ratios as in Figures 9 and 10 of Parrish et al. [1992]. The alkane present in the lowest concentrations and therefore subject to the greatest experimental uncertainties is placed in the numerator, and the most precisely determined ratio is used as the abscissa. Only data with/-butane concentrations greater than twice the detection limit are included in the plot, although the full data sets were used in the linear, least squares fits discussed below. ) The OH kinetics provide a mechanism for the evolution of the ratios characteristic of the urban emission source areas to the ratios observed in aged marine (Atlantic data) and continental (Boulder data) areas in Figure lb. These