Intercomparison of tropospheric OH and ancillary trace gas measurements at Fritz Peak Observatory, Colorado

Abstract

The determination of the concentration of OH in the Earth's troposphere is of fundamental importance to an understanding of the chemistry of the lower atmosphere. Although many experiments to measure OH concentration have been performed in recent years, very few operate at sensitivities necesssary to measure the extremely low amount of OH in the clean troposphere (0.1-0.2 parts per trillion by volume at summertime local noon). This paper describes an informal intercomparison campaign held at Fritz Peak, Colorado, in summer 1991 to intercompare the OH concentrations determined from a spectroscopic instrument and an in situ chemical conversion instrument, both with sensitivities at or below 5 x 105 molecules cm -3. Ancillary measurements including those of 0 3, CO, NO, NO 2, NO., H20, SO 2, aerosols, solar flux, and meteorological parameters were also performed to test photochemical theories of OH formation. These measurements also provided a means for comparing air masses at the long path and in situ sites. The intercomparison was very successful with measured values of OH concentration in agreement within one standard error much of the time. OH concentrations were typically low, rarely above 4 x 106 cm -3, with only slow growth during the morning hours, indicating the possible presence of scavenger species. Model results suggest higher than measured OH concentrations or the presence of scavenger species. Logan et al. [1981] have placed OH chemistry in the framework of the global troposphere. Tropospheric OH is formed (mainly) by the reactions EISELE ET AL.' TROPOSPHERIC OH AND TRACE GAS MEASUREMENTS 18,607 Boulder/Gilpin County, Colorado Caribou Mine ? er ,,,17 km •" '•'"'•':• N e d e r I a n d Eldore North o linsvi e contour intsrval-122 meters Fritz Peak Observatory 2,676m Figure 2. Topographic map of the area, including Fritz Peak Observatory and Caribou with local (very small) towns shown. Wind is typically from the west. measurement of atmospheric absorption of individual rotational lines in the ultraviolet electronic OH bands near 308 nm; (5) a two-channel arra• detector system with signalto-noise ratio greater than 10•* per integration that allows simultaneous observation of both laser and atmospheric absorbed laser light at all wavelengths in a spectral interval sufficient to include measurement of several OH rotational Caribou Mlnl• / Ob•,•tmO, , ,o.• / -I 2.50 m Retro Reflector Array Laser • • OH Path Geometry Cross Section As Seen From East concentration of H2SO 4 and methane sulfonic acid and estimates of H2SO 4 production and loss in the atmosphere, J. Geophys. Res., 98, 9001-9010, 1993. Evaluation of a catalytic reduction technique for the measurement of total reactive odd-nitrogen, NOy, in the atmosphere, J. Atmos. Chern., 3, 435-468, 1985. nitrogen species in the troposphere: Measurements of NO, NO 2, HNO 3, particulate nitrate, peroxyacetyl nitrate (PAN), 03, and total reactive odd nitrogen (NO) at Niwot Ridge, Colorado, J. Geophys. Res., 91, 9781-9793, 1986. Impact of natural hydrocarbons and hy...

Daytime HO2 concentrations at Oki Island, Japan, in summer 1998: Comparison between measurement and theory

“…Stevens et al [1997] performed another steady state model and found that HO2 is overestimated by a factor of 3-4 under clean conditions with NO of less than 100 parts per trillion by volume (pptv), suggesting that the slower rates of RO2 + NO reactions and faster rates of RO2 + HO2 reactions for RO2 radicals derived from higher hydrocarbons could reduce the discrepancies. More comprehensive steady state model calculations performed by McKeen et al [1997] have still overpredicted OH concentrations as they did in the 1991 intercomparison studies at a site in the same region [Eisele et al, 1994], even with the larger hydrocarbon data set. When some hydrocarbon compounds which behave similarly to [•-pinene were added to the model in order to decrease the modeled OH to the observed concentrations, total HOx + RO2 was also decreased and the factor of HO2 overprediction was reduced from 4 to 2 Recently, OH and HO2 concentrations observed at Mace Head (53ø20'N, 9ø54'W) during July and August 1996 were compared to model predictions [Carslaw et al, 1999a].…”

1993 Tropospheric OH Photochemistry Experiment: A summary and perspective

“…The OH concentrations measured in 1991 were much less than expected. A matching interpretation via a photochemical model required the addition of the equivalent of 2 ppb of isoprene (an unrealistically large amount of this compound alone) to the species responsible for OH removal [Eisele et al, 1994]. This signaled a serious deficiency in our understanding of OH loss rates in a Consequently, a major priority of the 1993 campaign was the quantification of OH loss rates through comprehensive measurement of nonmethane hydrocarbons.…”

“…For the 1991 measurements, model calculations [Eisele et al, 1994] used loss rates due to CO, CH4, and anthropogenic nonmethane hydrocarbons (the latter were not measured but scaled to observed NOx, although their loss rate contribution was minor). Predicted OH was 3 to 5 times measured values, and biogenically emitted hydrocarbon compounds that react with OH were thought the likely reason.…”

On the role of hydroxyl radicals in the self-cleansing capacity of the troposphere

“…Although OH can be measured quite accurately (Perner et al, 1987;Eisele et al, 1994;Hard et al, 1995;Brune et al, 1998;Heard and Pilling, 2003), its variability precludes that the spatially and temporally averaged OH-field, a measure for the atmospheric self-cleansing capacity, is determined from direct observations. An indirect method has therefore been developed, using long-term measurements of the tracer gas methyl chloroform (MCF), a compound which is largely removed from the atmosphere by OH and has only anthropogenic sources (Lovelock, 1977;Singh, 1977;Prinn et al, 1992).…”