Ozone fluxes in a Pinus ponderosa ecosystem are dominated by non-stomatal processes: Evidence from long-term continuous measurements

Abstract

Ecosystems remove ozone from the troposphere through both stomatal and non-stomatal depositions. The portion of ozone taken up through stomata has an oxidative effect causing damage. We used a multi-year dataset to assess ozone deposition to a ponderosa pine plantation near Blodgett Forest, Georgetown, California. Environmental parameters, water and ozone concentrations and fluxes were measured continuously from January 2001 to December 2006. High levels of ozone concentrations (up to 100 ppb) were observed during the spring–summer period, with corresponding high levels of ozone fluxes (up to 30 μmol m⁻² h⁻¹). During the summer season, we calculated that a large portion of the total ozone flux was due to non-stomatal processes, which is in agreement with previous studies suggesting that chemical reactions with BVOCs (biogenic volatile organic compounds) emitted by the ecosystem are mainly responsible for this ozone flux. We also report here the first direct measurement of BVOC + ozone oxidation products, confirming that ozone loss process is occurring below our flux measurement height. We analyzed the correlations of common ozone exposure metrics based on accumulation of concentrations (AOT40 and SUM0) with ozone fluxes (total, stomatal and non-stomatal). Stomatal flux, which is considered responsible for ozone damage, showed a weaker correlation with ozone concentrations than non-stomatal flux during summer and fall seasons. The non-stomatal flux is more strongly correlated with ozone concentration because BVOC emission and ozone concentration both increase with temperature. We suggest that AOT40 and SUM0 are poor predictors of stomatal ozone uptake, and that a physiologically based metric would be more effective.

Introduction

Ozone is considered one of the most dangerous oxidant molecules for plants (UNECE, 2004, US, 2007). Ozone concentration in the northern midlatitude atmosphere is increasing (Brasseur et al., 1998). It was estimated that by 2050 the average atmospheric concentration will exceed the 40 ppb threshold currently used for estimating oxidative damage to vegetation (Langner et al., 2005). Chronic stress by exposure to moderate ozone concentrations usually produces biochemical and physiological changes (Darrall, 1989, Sandermann et al., 1997, Zheng et al., 2002). The inhibition of carbon assimilation by photosynthesis and a decrease in plant growth is a common effect (Guderian et al., 1985), often associated with visible injuries (Bussotti et al., 2003, Vollenweider and Gunthardt-Goerg, 2005) under conditions of acute ozone concentrations in the atmosphere. Decline in stomatal conductance is a common effect of atmospheric ozone, so the capability of plants to exchange water and CO₂ will be negatively affected by rising ozone concentrations in the atmosphere (Wittig et al., 2007).

Plants act as a sink for ozone, through stomatal and non-stomatal processes. Stomatal conductance to ozone is the inverse of the sum of an array of resistances that ozone meets in specific locations along the path from outside the leaf to the reaction site inside the apoplast (Fares et al., 2008). Non-stomatal processes include ozone deposition to soil, stems, cuticles, and, in general, any external surface, along with gas-phase chemical losses involving reactions between ozone and BVOCs (biogenic volatile organic compounds) or nitric oxide (NO) emitted by the ecosystem (plants and soils) (Kurpius and Goldstein, 2003). In parallel with ozone formation, the emissions of some BVOCs increase with light (Niinemets et al., 2004) and exponentially with temperature (Tingey et al., 1991, Monson et al., 1992, Guenther et al., 1995), consequently producing higher emissions during spring–summer seasons (Holzinger et al., 2006) when ozone concentrations are highest.

At a regulatory level, identifying a reliable metric for ozone risk assessment is important. The metrics used for ecological ozone risk assessment are mostly based on the accumulated daytime ozone concentration. Typically, hourly O₃ concentrations from the NCLAN (National Crop Loss Assessment Network) program are used to calculate different means and cumulative statistics. Several metrics were developed using both threshold and functional concentration weighting and tested for best fit to NCLAN yield responses (Lefohn and Runneckles, 1987, Musselman et al., 1988, Rawlings et al., 1988) and described in EPA (US Environmental Protection Agency) directives (US, 1986, US, 1996). Concentration-based metrics calculated from 8 am to 8 pm were also used by EPA for generating a national exposure surface with the intent to analyze national ozone air quality, and develop maps of vegetation exposures and risk under different ozone level regimes and applied standards (EPA, 2007). These metrics are hourly averages of ozone concentration, SUM06 (sum of all ozone values greater than or equal to 0.06 parts per million) and W126 (a weighted sum of all ozone values).

The best representative metric in a ponderosa pine ecosystem in the Sierra Nevada mountains of California was found to be SUM0 (Panek et al., 2002, Kurpius and Goldstein, 2003), based on the sum of all daytime ozone concentrations. Therefore, we used the SUM0 metric in this study. The European directives (UNECE, 2004) suggest the use of the AOT40 (accumulated ozone over a threshold concentration of 40 ppb) for forest ecosystems, a metric which considers only daylight hours over a certain solar radiation intensity (Karenlampy and Skarby, 1996, Fuhrer et al., 1997). This second metric was also adopted in this study and compared to SUM0. Ozone concentration at the canopy-level is used to calculate these metrics.

Ozone concentration is not always correlated to ozone flux (Kurpius et al., 2002), and accumulated exposure to ozone does not take into account the physiology of plants nor the effective dose of ozone absorbed by plants via stomata (Panek et al., 2002). Drought stress in Mediterranean ecosystems induces stomatal closure, consequently limiting the ozone uptake by stomatal absorption even at high ozone concentrations (Emberson et al., 2007). This may result in much lower damage than predicted by the AOT40 index. In order to establish a cause–effect relationship considering only the effective concentration of ozone entering the leaf, a stomatal flux-based index has been considered (Karlsson et al., 2004, Simpson et al., 2007, Tuovinen et al., 2007, Tuovinen, 2009, Matyssek et al., 2007), and proposed for some tree species (Beech and Birch) in the risk assessment methodology adopted within the CLRTAP (Convention of Long-Range Transboundary Air Pollution) of UNECE (UNECE, 2004). Partitioning ozone flux between the stomatal and non-stomatal mechanisms is considered the best choice for ozone risk assessment (Emberson et al., 2000), although estimating the relative amount of the stomatal ozone flux depends on the ecophysiology of a specific ecosystem and requires more detailed information than is often available. Stomatal ozone uptake was found to be the major contributor to the total ozone flux at the whole plant and leaf level (Fredericksen et al., 1996, Fares et al., 2008), when adsorption sinks are limited and in dry conditions (Altimir et al., 2006). However, on wet plant surfaces and in presence of BVOC, considerable amounts of ozone can react with a multitude of waxes, salts, ions, and many other compounds leading the non-stomatal mechanisms to represent up to 30–70% of the total ozone flux in a Pinus ponderosa ecosystem (Kurpius and Goldstein, 2003, Goldstein et al., 2004), in a sitka spruce ecosystem (Coe et al., 1995), in a Mediterranean oak forest (Gerosa et al., 2005) in a northern mixed hardwood forest (Hogg et al., 2007) and in a sub-alpine ecosystem (Zeller and Nikolov, 2000).

In this work, we illustrate the dynamics of stomatal and non-stomatal ozone fluxes measured from 2001 to 2006 in a ponderosa pine plantation located in the Sierra Nevada Mountains of California. The goal of this study was (1) to examine the longer term dataset for consistency with previous shorter term studies performed in the same forest ecosystem at an earlier stage of development which suggested that non-stomatal ozone fluxes were a dominant portion of the total ozone fluxes; (2) to determine whether plant physiology dominates both stomatal and non-stomatal ozone fluxes (3) to show how well ozone fluxes correlate with ozone concentrations during the four seasons, through the use of two common metrics (SUM0 and AOT40) correlated with the accumulated total, stomatal and non-stomatal ozone fluxes.