Investigation of Mercury Exchange between Forest Canopy Vegetation and the Atmosphere Using a New Dynamic Chamber † JENNIFER A. GRAYDON,* ,‡ VINCENT L. ST. LOUIS, ‡ STEVE E. LINDBERG, § HOLGER HINTELMANN, | AND DAVID P. KRABBENHOFT ⊥ Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada, T6G 2E9, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, Tennessee 37831-6083, Department of Chemistry, Trent University, 1600 West Bank Drive, Peterborough, Ontario, Canada, K9J 7B8, and U.S. Geological Survey-WRD, 8505 Research Way, Middleton, Wisconsin 53562 This paper presents the design of a dynamic chamber system that allows full transmission of PAR and UV radiation and permits enclosed intact foliage to maintain normal physiological function while Hg(0) flux rates are quantified in the field. Black spruce and jack pine foliage both emitted and absorbed Hg(0), exhibiting compensation points near atmospheric Hg(0) concentrations of ∼2-3 ng m -3 . Using enriched stable Hg isotope spikes, patterns of spike Hg(II) retention on foliage were investigated. Hg(0) evasion rates from foliage were simultaneously measured using the chamber to determine if the decline of foliar spike Hg(II) concentrations over time could be explained by the photoreduction and re-emission of spike Hg to the atmosphere. This mass balance approach suggested that spike Hg(0) fluxes alone could not account for the measured decrease in spike Hg(II) on foliage following application, implying that either the chamber underestimates the true photoreduction of Hg(II) to Hg(0) on foliage, or other mechanisms of Hg(II) loss from foliage, such as cuticle weathering, are in effect. The radiation spectrum responsible for the photoreduction of newly deposited Hg(II) on foliage was also investigated. Our spike experiments suggest that some of the Hg(II) in wet deposition retained by the forest canopy may be rapidly photoreduced to Hg(0) and re-emitted back to the atmosphere, while another portion may be retained by foliage at the end of the growing season, with some being deposited in litterfall. This finding has implications for the estimation of Hg dry deposition based on throughfall and litterfall fluxes. Introduction Numerous studies have shown that the forest canopy con- tributes significantly to fluxes of monomethylmercury (MMHg) and total Hg (THg; all forms of Hg in a sample) to watersheds (1-5). For example, St. Louis et al. (6) found that annual fluxes of MMHg and THg in throughfall (wet depo- sition that passes through the forest canopy) plus litterfall at the Experimental Lakes Area (ELA) in the Boreal ecoregion of northwestern Ontario, Canada, were approximately two and three times greater, respectively, than fluxes of MMHg (0.1 µgm -2 ) and THg (7 µgm -2 ) in wet deposition over open areas such as lakes. Almost all the increased flux of Hg under the forest canopy occurred as litterfall, and, as a result, it is important to understand whether litterfall constitutes a new and/or recycled input of Hg to the forest floor (1, 6-7). For litterfall to be considered a new input, the Hg in/on foliage must originate from the atmosphere. For example, atmo- spheric particulate Hg (pHg) and reactive gaseous Hg (RGM) can be dry deposited to foliar surfaces (8) and atmospheric Hg(0) can be assimilated into foliage through stomata (9-11). In addition to the uptake of new Hg(0) from the atmosphere, Hg(0) released from soils below the canopy through volatilization could also be taken up via stomata, resulting in Hg being recycled within the forest. Plants may also accumulate Hg from soil waters by root uptake and translocation to foliage (12), although previous studies have shown that there is generally little accumulation by foliage via this pathway except in areas where soil Hg content is high (9). Plant foliage also provides an excellent surface for photochemical reduction of newly deposited Hg(II) remain- ing in the forest canopy following precipitation events, affecting net deposition of Hg(II) to the forest floor (13). Both micrometeorological and chamber techniques have been used to measure exchange of Hg(0) at the biosphere/ atmosphere interface, but most previous studies have been conducted in areas either naturally enriched or anthropo- genically contaminated with Hg (14-16). Unlike microme- teorological techniques, chamber methods utilize small plot sizes, making it possible to compare responses of enclosed vegetation to experimental treatments (17). However, the microclimate of vegetation enclosed in chambers is invariably altered from natural conditions, with the degree of alteration depending on the design of the chamber (18). Dynamic chambers, where there is a net flow of air through the chamber, are the most common design used to quantify gas exchange. Several studies have shown that low air turnover rate in dynamic chambers can result in artificially large boundary layers being established on soil/foliar surfaces, resulting in underestimates of true Hg(0) emissions (e.g., 19). Furthermore, since there is evidence that a significant amount of foliar Hg(0) exchange occurs via stomata (2, 8, 10), it is also critical that chamber designs do not interfere with normal physiological responses (e.g., photosynthesis and transpiration) of enclosed foliage. Materials used in the construction of chambers can also have a profound influence on the effectiveness of the system to measure gas exchange (18, 20). This is especially true for chambers designed to measure Hg(0) exchange, because in addition to acting as a potential source of Hg(0) contamination, some chamber materials may also adsorb/absorb Hg(0), or block spectral wavelengths such as ultraviolet (UV) radiation important for the photochemical reduction of Hg(II) to Hg(0) (21). Here, we present the design of a new dynamic chamber system that allows both full spectrum radiation transmission and enclosed foliage to maintain normal physiological function while Hg(0) flux rates are quantified. Using this Hg(0) flux chamber in conjunction with isotopically enriched Hg(II) spikes, we examined the following: (1) ambient Hg(0) exchange with foliage over a small natural range of atmo- * Corresponding author phone: (780) 492-0900; fax: (780) 492- 9234; e-mail: jgraydon@ualberta.ca. † Contribution No.14 of the Mercury Experiment to Assess Atmospheric Loading in Canada and the U. S. (METAALICUS). ‡ University of Alberta. § Oak Ridge National Laboratory. | Trent University. ⊥ U.S. Geological Survey-WRD. Environ. Sci. Technol. 2006, 40, 4680-4688 4680 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 15, 2006 10.1021/es0604616 CCC: $33.50  2006 American Chemical Society Published on Web 07/04/2006