Missing OH Reactivity in a Forest: Evidence for Unknown Reactive Biogenic VOCs

A. Guentheret al., J. Geophys. Res.100, 8873 (1995).

M. Traineret al., Nature329, 705 (1987).

W. L. Chameides, R. W. Lindsay, J. Richardson, C. S. Kians, Science241, 1473 (1988).

J. D. Fuenteset al., Bull. Am. Meteorol. Soc.81, 1537 (2000).

L. I. Kleinmanet al., Geophys. Res. Lett.29, 10.1029/2001GL014569 (2002).

T. B. Ryersonet al., Science292, 719 (2001).

S. E. Paulson, A. D. Sen, P. Liu, J. D. Fenske, M. J. Fox, Geophys. Res. Lett.24, 3193 (1997).

N. M. Donahue, J. H. Kroll, J. G. Anderson, K. L. Demerjian, Geophys. Res. Lett.25, 59 (1998).

T. Pfeiffer, O. Forberich, F. J. Comes, Chem. Phys. Lett.298, 351 (1998).

M. O. Andreae, P. J. Crutzen, Science276, 1052 (1997).

Z. Meng, D. Dabdub, J. H. Seinfeld, Science277, 116 (1997).

C. D. O'Dowd, P. Aalto, K. Hämeri, M. Kulmala, T. Hoffmann, Nature416, 497 (2002).

I. G. Kavouras, N. Mihalopoulos, E. G. Stephanou, Nature395, 683 (1998).

R. J. Griffin, D. R. Crocker, R. C. Flagan, J. H. Seinfeld, J. Geophys. Res.104, 3555 (1999).

I. Faloonaet al., J. Geophys. Res.106, 24315 (2001).

M. R. Kurpius, A. H. Goldstein, Geophys. Res. Lett., 30, 1371, 10.1029/2002GL016785 (2003).

M. A. Carroll, S. B. Bertman, P. B. Shepson, J. Geophys. Res.106, 24275 (2001).

T. A. Kovacs, W. H. Brune, J. Atmos. Chem.39, 105 (2001).

T. A. Kovacset al., J. Environ. Monit.5, 68 (2003).

N. M. Donahue, J. S. Clarke, K. L. Demerjian, J. G. Anderson, J. Phys. Chem.100, 5821 (1996).

F. Kaufmann, J. Phys. Chem.88, 4909 (1984).

P. S. Stevens, J. H. Mather, W. H. Brune, J. Geophys. Res.99, 3543 (1994).

Each decay takes 4.3 min, with 20 s at each of 13 steps, a 10-s measurement of OH plus the background signal, and a 10-s measurement of the background signal alone. The OH signal decreases a factor of 10 to 20 over the 13 steps. The OH reactivity, k_OH, is the slope of the logarithm of the OH signal, S_OH, as a function of the time (the distance divided by the velocity) minus the OH loss to the flow tube's walls, k_wall: k_OH = –Δln(S_OH)/Δtime – k_wall. The OH wall loss, which was determined by flowing ultrahigh-purity zero air down the flow tube, was 2.2 ± 0.4 s^–1 for PROPHET 2000 and five subsequent field studies. We calibrated the TOHLM by adding known quantities of CO and some hydrocarbons and comparing the slope of the OH reactivity with the accepted reaction-rate coefficient (18). In the presence of ambient NO, HO₂ reacts with the NO to reform OH, changing the observed slope of the decay. The HO₂ recycling was insignificant during the PROPHET 2000 campaign, when NO was never more than 0.2 ppbv.

W. R. Stockwell, F. Kirchner, M. Kuhn, J. Geophys. Res.102, 25847 (1997).

The absolute uncertainty of TOHLM measurement is ±10%, with 1σ confidence level, and the wall loss uncertainty is ±0.4 s^–1. For an OH reactivity of 10 s^–1, the combined uncertainty is ±1.05 s^–1, with 1σ confidence level. The total uncertainty in the calculated OH reactivity [14% on average (1σ confidence level)] is based on the uncorrelated errors of the measurements of the atmospheric constituents and their reaction-rate coefficients and the SEM for each temperature (19). The gas chromatography–flame ionization detector measurements of V.Y. and co-workers in 2000, using the Al₂O₃/KCl porous layer open tubular (PLOT) column, indicate no significant hydrocarbon-like molecules were unaccounted for. In single-ion gas chromatography–mass spectrometry measurements that were tuned to measure monoterpenes, made by D.R. and E.A. in 1998, the peaks of all VOCs in the monoterpene region were identified and quantified above an equivalent mixing ratio of 1 pptv; less than 1% of the total peak area was unidentified (17).

G. W. Schade, A. H. Goldstein, M. S. Lamanna, Geophys. Res. Lett.26, 2187 (1999).

D. Tanet al., J. Geophys. Res.106, 24407 (2001).

R. Atkinson, J. Arey, Atmos. Environ.37 (suppl 2), 197 (2003).

The ratio of the reaction rate of the constituent with OH to the product of the reaction rate with O₃ and its OH yield is given by k_OH/(k_O3·Y) = (OHreac_meas – OHreac_calc)/{OHloss – OHsource_calc)/[O₃]}, where OHreac_meas is the OH reactivity measured; OHreac_calc is the OH reactivity calculated (the difference is the missing OH reactivity); OHloss is the OH loss rate, the product of the OH concentration and the measured OH reactivity; OHsource_calc is the OH production from all known OH sources, both photolytic sources and those involving O₃ plus known alkenes; and [O₃] is the measured concentration of ozone.

R. Atkinson, J. Phys. Chem. Ref. Data26, 215 (1997).

S. M. Ashmann, J. Arey, R. Atkinson, Atmos. Environ.36, 4347 (2002).

A. C. Lewiset al., Nature405, 778 (2000).

X. Renet al., Atmos. Environ.37, 3639 (2003).

We thank J. Bassis and J. Simpas, G. Yip for assistance with the CO data, G. Visconti for support, A. D'Altorio for helpful discussions, the University of Michigan Biological Station for their logistical support of PROPHET 1998 and 2000, and the Atmospheric Chemistry Division of the National Center for Atmospheric Research for the loan of tower components. Supported by grants from NSF (no. ATM-9974335) for measurements at Prophet and from NOAA (no. 40RANROM1560) for measurements in Houston (W.H.B.), and by additional funding from NSF to the University of Michigan (M.A.C.), Purdue University (P.B.S.), and Ohio University (V.L.Y.); from the National Center for Atmospheric Research (E.A. and D.R.); and by the Italian Agency of Space and Center of Excellence on Remote Sensing and Numerical Modeling for the Forecast of Severe Weather (CETEMPS) (P.D.C.).