Anthropogenic, biogenic, and photochemical influences on surface formaldehyde and its significant decadal (2006-2017) decrease in the Lewiston-Clarkston valley of the northwestern United States.

Formaldehyde (HCHO) is a key carcinogen and plays an important role in atmospheric chemistry. Both field measurements and Positive Matrix Factorization (PMF) modeling have been employed to investigate the concentrations and sources of HCHO in the Lewiston-Clarkston (LC) valley of the mountainous northwestern U.S. Different instruments were deployed to measure surface formaldehyde and other related compounds in July of 2016 and 2017. The measurements reveal that the average HCHO concentrations have significantly decreased to 2-5 ppb in the LC valley in comparison to its levels (10-20 ppb) observed in July 2006. This discovery with surface measurements deserves attention given that satellite retrievals showed an increasing long-term trend from 2005 to 2014 in total vertical column density of HCHO in the region, suggesting that satellite instruments may not adequately resolve small valleys in the mountainous region. Our PMF modeling identified four major sources of HCHO in the valley: (1) emissions from a local paper mill, (2) secondary formation and background, (3) biogenic sources, and (4) traffic. This study reveals that the emissions from the paper mill cause high HCHO spikes (6-19 ppb) in the early morning. It is found that biogenic volatile organic compounds (VOCs) in the area are influenced by national forests surrounding the region (e.g., Nez Perce-Clearwater, Umatilla, Wallowa-Whitman, and Idaho Panhandle National Forests). The results provide useful information for developing strategies to control HCHO levels and have implications for future HCHO studies in atmospheric chemistry, which affects secondary aerosols and ozone formation.

Comparison of ultraviolet absorbance, chemiluminescence, and DOAS instruments for ambient ozone monitoring.

This paper evaluates the accuracy of ozone measurements made by monitors that determine ozone concentrations in ambient air by UV absorption. These monitors are typically used to measure ozone for the purpose of establishing local compliance to air-quality standards. The study was predicated by the concern that commercially available UV absorbance O3 monitors may be subject to interference from volatile organic carbon (VOC) species that absorb light at 254 nm. To test for these and other effects, we compared simultaneous O3 measurements made by a commercial UV O3 monitor with an O3-NO chemiluminescence instrument, which is not subject to interference by VOC compounds. The comparisons were carried out in the summers of 1999 and 2000 at urban/industrial sites in Nashville and Houston, and in 2004 aboard a ship in the Gulf of Maine. In the two urban areas, we also compared the 03 measurements from these two methods with O3 measurements made by a long-path differential optical absorption spectrometer (DOAS). Our tests indicate that, with well-maintained monitors, there are no significant interferences even in areas with significant ambient concentrations of potentially interfering VOCs.

Intercomparison of volatile organic carbon measurement techniques and data at La Porte during the TexAQS2000 Air Quality Study.

The Texas Air Quality Study 2000 (TexAQS2000) investigated the photochemical production of ozone and the chemistry of related precursors and reaction products in the vicinity of Houston, TX. The colocation of four instruments for the measurement of volatile organic carbon compounds (VOCs) allowed a unique opportunity for the intercomparison of the different in-situ measuring techniques. The instruments included three gas chromatographs, each with a different type of detector, and a Proton-Transfer-Reaction Mass Spectrometer (PTR-MS) with each system designed to measure a different suite of VOCs. Correlation plots and correlation statistics are presented for species measured by more than one of these instruments. The GC instruments were all in agreement to within 10-20% (slope) with coefficients of variation (r2) of > or = 0.85. The PTR-MS agreement with other instruments was more dependent on species with some very good agreements (r2 values of approximately 0.95 for some aromatics), but isoprene, acetaldehyde and propene were substantially less highly correlated (0.55 < r2 < 0.80). At least part of these differences were undoubtedly due to the timing of sample acquisition in an environment in which VOC levels changed very rapidly on both quantitative and temporal scales.