Northeastern Naturalist Vol. 21, No. 4 L.K. Seiler, D.R. Decoteau, and D.D. Davis 2014 541 2014 NORTHEASTERN NATURALIST 21(4):541–553 Evaluation of Ailanthus altissima as a Bioindicator to Detect Phytotoxic Levels of Ozone Lauren K. Seiler1, Dennis R. Decoteau2, and Donald D. Davis3,* Abstract - Ozone is a common ground-level air pollutant that can be harmful to human health, vegetation, and ecosystems. Ailanthus altissima (Tree-of-Heaven) is an exotic, invasive tree species found across most of the US that has been suggested as a bioindicator to detect phytotoxic levels of ambient ozone. We conducted greenhouse exposures and field studies to evaluate the ozone-sensitivity of Tree-of-Heaven, including seedlings from several seed sources. Severity of ozone-induced foliar symptoms was directly related to ozone concentration and varied among seed sources. Tree-of-Heaven seedlings out-planted in the field developed foliar symptoms induced by ambient ozone levels at or near the national ambient air quality standard (NAAQS) for photochemical oxidants and ozone. Our results provide data that may be considered when setting new a NAAQS for photochemical oxidants and ozone. Introduction Ground-level ozone is harmful to human health, vegetation, and ecosystems and is responsible for $500 million in reduced annual crop yield in the US (USEPA 2003). Ozone stress is especially severe within forested ecosystems in the northeastern US and across Pennsylvania (PA) (Orendovici-Best et al. 2010, Skelly 2000). Monitoring ambient levels of ozone in this area and documenting its effects on vegetation may prove useful in setting federal national ambient air quality standards (NAAQS) for photochemical oxidants and ozone (USEPA 2006, 2007, 2010). Our laboratory annually monitors ambient levels of ground-level ozone at four rural locations in PA. Monitoring results indicate that phytotoxic concentrations of ozone occur annually in forested regions in the state especially at the higher-elevation, mountainous sites in the more remote north-central region (Orendovici-Best et al. 2010). Native forests in north-central PA may be at risk from ambient levels of ozone, despite their perceived pristine location. In addition, we have conducted field surveys to document the presence or absence of ozone-induced symptoms on native plants (Davis 2007a, b; 2009; 2011). However, we currently lack a common, widespread, perennial ozone-sensitive bioindicator plant that grows on a variety of sites to further document the occurrence of phytotoxic levels of ambient ozone in remote forested areas. 1Ecology Program, Penn State Institutes of Energy and the Environment, The Pennsylvania State University, University Park, PA 16802. 2Department of Plant Science Penn State Institutes of Energy and the Environment, The Pennsylvania State University, University Park, PA 16802. 3Department of Plant Pathology and Environmental Microbiology, Penn State Institutes of Energy and the Environment, The Pennsylvania State University, University Park, PA 16802. *Corresponding author - ddd2@psu.edu. Manuscript Editor: David Richardson Northeastern Naturalist 542 L.K. Seiler, D.R. Decoteau, and D.D. Davis 2014 Vol. 21, No. 4 Manning and Federer (1980) advocated the use of bioindicators to yield a general indication of presence or absence of phytotoxic levels of ambient ozone. Bioindicators offer an alternative to the use of chemical/mechanical air monitors that directly measure ambient pollutant levels. Although plant bioindicators do not yield quantitative air-quality data for specific pollutants, they have advantages over mechanical/chemical means, including lower costs, less maintenance, and greater flexibility with respect to expansion or relocation. In addition, because they are part of the ecosystem, bioindicators are likely to be more directly correlated with the health of the ecosystem than are human-made monitors. Since the discovery that leaf stipple on Vitis sp. (grape) was caused by ozone (Richards et al. 1958), adaxial dark stipple has been the classic foliar symptom used to evaluate levels of ozone-induced symptoms on broadleaved bioindicators (Skelly 2000, Skelly et al. 1987). The presence of stippling has been utilized as the standard broadleaf-plant response to ozone in many laboratory, greenhouse, and field studies (Skelly 2000). In 1974, we reported that Tree-of-Heaven was sensitive to ozone within a controlled environment chamber in our laboratory, with 74% of 35 exposed seedlings developing dark stipples on the adaxial leaf surface, as well as adaxial tan stipples, chlorosis, and premature defoliation (Davis and Coppolino 1974). In addition to documented effects on seedlings, injury due to ambient ozone has been observed on Tree-of-Heaven saplings and trees watered only by natural rainfall in various areas in PA (D.D. Davis, pers. observ), as well as within a wildlife refuge in New Jersey (Davis and Orendovici 2006). The predominant ozone-induced foliar symptom on Tree-of-Heaven in these surveys was a dark adaxial stipple, but the species also occasionally exhibited only adaxial tan stipple. Based on these observations, we proposed that Tree-of-Heaven might be a useful ozone-sensitive bioindicator (USDOI 2003). Likewise, Gravano et al. (2003) reported that potted, watered Tree-of-Heaven root suckers were useful bioindicators in Italy, where ambient ozone induced a dark adaxial leaf stipple. However, to the best of our knowledge, no studies have been conducted under a combination of carefully controlled exposures and field observations that included rigorous statistical analyses. The main objective of this study was to more thoroughly evaluate the usefulness of Tree-of-Heaven as an ozone-sensitive bioindicator. Methods We used two experimental approaches to evaluate the efficacy of Tree-of-Heaven as a bioindicator of ozone pollution. First, potted Tree-of-Heaven seedlings were exposed to ozone within continuously stirred tank reactor (CSTR) chambers (Heck et al. 1975) in a greenhouse during 2010 and 2011 to establish the relationship between controlled ozone-exposure concentrations and induced foliar stipple and to determine if response to ozone varied among seedlings originating from several seed sources across the country. Second, we planted Tree-of-Heaven seedlings in a central PA field in 2011 to observe the species’ response to ambient ozone, as measured at two nearby Environmental Protection Agency (EPA) monitoring sites. Northeastern Naturalist Vol. 21, No. 4 L.K. Seiler, D.R. Decoteau, and D.D. Davis 2014 543 2010 greenhouse exposures As part of a related study (Kasson et al. 2013), we acquired Tree-of-Heaven seed from six locations across the US: Bloomington, IN; Corvallis, OR; Far Rockaway, NY; Mineral, VA; Reno, NV; and State College, PA (M.T. Kasson, Penn State, State College, PA, pers. comm.). We placed one hundred Tree-of-Heaven seeds from each source in a refrigerator maintained at approximately 2 °C. Two weeks later, we placed the seeds on flats containing a general potting mix, moved them to a greenhouse, and watered all flats. The greenhouse, located on The Pennsylvania State University campus, had been modified to pass incoming air through activated-charcoal filters, which reduced the ambient ozone concentration within the greenhouse to less than 8 ppb ozone. Upon germination, we planted 64 Tree-of-Heaven seedlings from each seed source in the greenhouse in individual 2-L pots containing SUN GRO Metro- Mix 200 growing medium (Sun Gro Horticulture Canada Ltd., Bellevue, WA) and fertilized them with ~5g Osmocote Plus® (15N:9P:12K) fertilizer (The Scotts Company LLC, Marysville, OH). We randomly arranged potted seedlings on greenhouse benches, where they received charcoal-filtered air, and watered them as needed. Prior to ozone exposure, four seedlings from each seed source were randomly assigned to one of the 16 CSTR chambers and to a specific location within the chamber. We initiated ozone exposure on 9 August 2010. Although all Tree-of-Heaven seedlings were of similar age (~6–7 weeks), height varied among seedlings originating from the various seed sources. Mean initial height (cm) of seedlings from each seed source was 22.2 (NY), 18.3 (NV), 17.9 (VA), 17.6 (PA), 13.9 (IN), and 12.1 (OR). Potted Tree-of-Heaven seedlings were exposed to 75 or 120 ppb ozone in a square-wave exposure for 8 hr/day, 5 days/week (Monday–Friday) for 3 weeks. Daily exposures began at 0900 hr and ended at 1700 hr. Seven of the 16 CSTR chambers were maintained at a target concentration of 75 ppb ozone, which was near the NAAQS for ozone (USEPA 2006), and seven chambers were set at a target of 120 ppb, a higher level intended to induce foliar symptoms in the event that conditions were not conducive for symptom development. Two control chambers received the ambient greenhouse charcoal-filtered air (less than 8 ppb ozone daily hourly average). We monitored ozone concentrations (ppb) in each chamber for 1.5 min at 12-min intervals during each exposure and sampled ozone through Teflon tubing using a solenoid-driven sampling system connected to a calibrated TECO Model 49 photometric ozone analyzer (Thermo Environmental Corp., Franklin, MA). Ozone and environmental data were input to a data-logger connected to a PC computer. During non-exposure hours, all plants remained in the CSTR chambers with the chamber doors open and were exposed to internal greenhouse environmental conditions, including charcoal-filtered air. Mean relative humidity was ~72% and mean temperature was ~29 oC in the chambers. We performed routine quality-control measures on monitoring equipment throughout the study. We visually rated Tree-of-Heaven leaves for ozone-induced symptoms at the end of each exposure week (on Sunday), using a modified Horsfall-Barratt classification scale (Horsfall and Barratt 1945). We rated each plant from 0–5 for amount (AMT) and severity (SEV) of ozone-induced foliar symptoms. The rating values Northeastern Naturalist 544 L.K. Seiler, D.R. Decoteau, and D.D. Davis 2014 Vol. 21, No. 4 corresponded to the following percentages: 0 = 0%, 1 = 1–6%, 2 = 7–25%, 3 = 26– 50%, 4 = 51–75%, and 5 = 76–100%. AMT described percentage of symptomatic leaves on a plant, whereas SEV described percentage of adaxial leaf surface that was symptomatic on affected leaves. We also made subjective observations, such as color of symptoms, but we did not include them in our statistical analyses. We converted AMT and SEV ratings for each plant to percentage values representing the midpoint of each injury class as follows: 0 = 0%, 1 = 3.5%, 2 = 16%, 3 = 38%, 4 = 63%, and 5 = 88%. We used AMT*SEV to represent whole plant injury, INJP. Mean INJP values for seedlings from each seed source, INJSS, were calculated and transformed using a log10 data transformation to attain equal variances. We employed a 2-way ANOVA to test for significant differences (P = 0.05) among transformed final (week 3) injury values for each seed source per treatment, and a Tukey HSD test to identify the significant differences (P = 0.05) (SAS 2011). 2011 greenhouse exposures In 2011, we followed the same protocol as in 2010. However, initial results of the 2010 study prompted several improvements in the 2011 methodology. In 2010, the potted Tree-of-Heaven seedlings from the OR seed source were more sensitive to ozone than seedlings from other seed sources. Therefore, to increase the possibility that the ozone would induce visible symptoms, we exposed only seedlings from the OR seed source in 2011. In addition, at the conclusion of the 2010 study, some CSTR chambers (e.g., on the west side of the greenhouse) appeared to be slightly warmer than others. Therefore, we monitored the pre-exposure temperature in all 16 chambers for 24 hr/day for 10 days prior to start of the 2011 exposures. We used the overall 10-day mean temperature for each chamber to group the exposure chambers into three blocks (five chambers/block) designated as low (mean = 24.6 ºC), medium (mean = 25.2 ºC), and high (mean = 25.8 ºC) temperature regimes. The 2010 results also suggested that measuring more ozone levels would allow better delineation of the relationship between ozone concentration and plant injury. Therefore, we used five target ozone concentrations (40, 60, 80, 100, and 120 ppb) in the CSTR chambers during 2011. Within each temperature block, we randomly assigned each individual chamber to one of the five target ozone-concentrations. One chamber served as the ozone control (less than 8 ppb ozone). We initiated ozone exposures at 0900 h on 11 July 2011, at which time seedlings were ~9 weeks old, and conducted the treatments 8 hrs/day, 5 d/wk (Monday–Friday) for 4 weeks. During exposures, mean relative humidity was ~80% and mean temperature was 25–26 ºC, as described above. Each Sunday, we rated the incidence and severity of ozone injury on each plant, as described for the 2010 studies. We converted ozone-induced foliar-injury ratings into percentages, and calculated AMT * SEV for each plant to represent whole-plant injury (INJP). A multiple regression analysis was performed using the mean final (week 4) INJP values, mean ozone concentration, mean initial plant height (cm), and mean temperature (ºC) for each chamber. Two interaction terms were used in the regression analysis: temperature x ozone (teoz) and height x ozone (heoz). We used SAS statistical software (SAS 2011) to analyze our data. Northeastern Naturalist Vol. 21, No. 4 L.K. Seiler, D.R. Decoteau, and D.D. Davis 2014 545 2011 field studies We germinated and grew Tree-of-Heaven seeds from one seed source (OR) in the greenhouse as in 2010. On 13 June 2011, we planted the thirty-six resultant seedlings (mean height = 24 cm) in a mowed field at Penn State’s Russell E. Larson Agricultural Research Center, Rock Springs, PA, located ~16 km west of the CSTR greenhouse facility. During dry weather, we watered seedlings as needed. We made weekly AMT and SEV ratings for foliar injury induced by ambient ozone on the Tree-of-Heaven seedlings for 9 weeks (18 July–13 September 2011), converted the ozone-induced foliar injury ratings into percentages, and calculated AMT*SEV for each plant to represent whole-plant injury (INJP). We downloaded hourly ambient ozone concentrations for the 2011 growing season from EPA AIRS site #42-027-9991, located ~1.1 km northeast of the field site. We calculated two indices, SUM60 and W126, from the downloaded ozone datasets and selected them as metrics of ambient ozone levels to determine if ambient ozone levels were related to the level of ozone injury on the bioindicator plants. SUM60 is the sum of mean hourly ozone averages ≥60 ppb, expressed in ppb.hrs. We selected this index because ambient SUM60 ozone has been correlated with ozone injury within eastern US forests (Hildebrand et al. 1996). We calculated the W126 index by multiplying hourly ozone averages between 0800 and 2000 hrs as a sigmoidal weighted function 1/(1 + 4403 * e[-126 * C]), where C is the hourly ozone concentration in ppb, which is added cumulatively over the growing season. The sum of these values (expressed in ppb.hrs) during the greatest consecutive 3-month period averaged over 3 years is calculated to determine the W126 ozone index. We selected the W126 index because the EPA has considered this index in calculations of a national secondary NAAQS for ambient ground-level ozone (USEPA 2007, 2010). Results 2010 greenhouse exposures The two-way ANOVA indicated that ozone concentration and seed source had significant (P < 0.0001) effects on plant injury in 2010. Tukey’s test (P = 0.05) results revealed that seedlings from the OR seed source exhibited significantly greater INJss values than seedlings from the other sources. Injury values for the other sources were statistically similar, with the exception of seedlings from the VA source, which exhibited significantly greater ozone injury than those from the NY seed source (Table 1). Seedlings from the NY source had the lowest INJss values of all seed sources, but injury values were statistically different only from the OR and VA seed sources. Control plants remained asymptomatic throughout the study. In both the 75 and 120 ppb treatments during 2010, seedlings from all seed sources except NY exhibited ozone injury after only 1 week of exposure to ozone (Fig. 1). At 75 ppb ozone, mean INJss was directly related to exposure duration for all six seed sources, but only seedlings from the OR seed source exhibited INJss values that were significantly different from the other seed sources. At 120 ppb ozone, mean INJss values were greater than those at 75 ppb. Injury levels increased with exposure duration at 120 ppb, except for seedlings from the VA seed source, Northeastern Naturalist 546 L.K. Seiler, D.R. Decoteau, and D.D. Davis 2014 Vol. 21, No. 4 which exhibited significantly greater INJss values at week 2, as compared to all seed sources except OR. Over the duration of both exposure concentrations, seedlings from the OR seed source exhibited the greatest INJss values, and seedlings from the NY source displayed the lowest values (Table 1, Fig. 1). 2011 greenhouse exposures Tree-of-Heaven seedlings in the control chamber, as well as those exposed to 40 ppb ozone, did not exhibit symptoms during the 2011 studies (data not shown). At 60–120 ppb ozone, the level of foliar injury was directly related to concentration and duration of ozone exposure (Fig. 2). Seedlings exposed to 60 ppb ozone did not exhibit foliar injury symptoms during weeks 1–2, but one of 12 plants (8.3%) exhibited foliar injury following weeks 3 and 4. Seedlings in the 80 ppb ozone treatment exhibited slight symptoms during weeks 1–3 of exposure, and greater injury at week 4. Plants in the 100 ppb ozone treatment exhibited levels of foliar-injury symptoms similar to those in the 80 ppb treatment during the first week of exposure, and symptoms then increased until week 3, at which time they leveled off. The temporal pattern of symptoms in the 120 ppb treatment was generally similar to those in the 100 ppb treatment, but more severe. Results of multiple regression analysis revealed the following significant (P = 0.0006, R2 = 0.8543) relationship between independent variables and final injury: (INJP = -34992 + [424 * temp] + [705 * ozone] + [76 * height] - [8 * teoz] - [2 * heoz]). The influence of each variable was significant in the regression equation. Ozone concentration had the greatest influence on INJP, followed by teoz, heoz, temperature, and height, respectively. 2011 field studies Tree-of-Heaven seedlings planted in the field displayed ozone-induced foliar injury symptoms when exposed to ambient ozone concentrations (Fig. 3). Seedlings displayed black stipple in the beginning of the monitoring period, but black stipple transitioned to tan stipple on 31 of 36 plants (86%) by the conclusion of the field survey. Tree-of-Heaven first exhibited ozone injury on 8 August. The level of injury increased in a non-linear fashion for the duration of the survey, except during 15–22 August, when ratings decreased. On the last day of evaluation, the final mean plantinjury value (INJP) was 177.5. Table 1. Seed sources and number of Tree-of-Heaven seedlings exposed to 75 and 120 ppb ozone in 2010. Mean final INJSS ± SE values represent average ozone-induced foliar injury per seed source for all ozone treatments at the end of the exposure period (week 3). Means followed by the same letter are not significantly different (P = 0.05) according to Tukey’s test. Seed source Number of plants Mean final INJSSA ± SE Corvallis, OR 64 1674.68 ± 215.61 c Mineral, VA 64 437.92 ± 88.44 b Reno, NV 64 359.15 ± 73.20 ab Bloomington, IN 64 311.08 ± 84.32 ab State College, PA 63 298.51 ± 58.44 ab Far Rockaway, NY 61 128.80 ± 35.78 a Northeastern Naturalist Vol. 21, No. 4 L.K. Seiler, D.R. Decoteau, and D.D. Davis 2014 547 Ambient levels of ozone first increased to 60 ppb (the first value to be included in SUM60 considerations) on 11 May (Fig. 3). SUM60 values continued to increase via a series of high ozone events over the course of the field study. The highest ozone-concentration events occurred during 8–10 June and 17–23 July. Figure 1. Mean foliar injury (INJSS ± SE) on Tree-of-Heaven seedlings from six seed sources following exposure to 75 ppb (upper graph) or 120 ppb (lower graph) ozone for 1, 2, or 3 weeks in 2010. Ozone-induced foliar injury was not observed on seedlings in the control (less than 8 ppb ozone). Northeastern Naturalist 548 L.K. Seiler, D.R. Decoteau, and D.D. Davis 2014 Vol. 21, No. 4 Figure 2. Mean foliar injury (INJP ± SE) on Tree-of-Heaven seedlings from the OR seed source exposed to 60, 80, 100, or 120 ppb ozone after 1, 2, 3, or 4 weeks in 2011. Ozoneinduced foliar injury was not observed on seedlings in the control (less than 8 ppb ozone) or 40 ppb ozone treatment. Figure 3. Cumulative ozone concentrations of 60 ppb or greater for the 2011 growing season (SUM60), cumulative weighted hourly ozone concentrations from 0800–2000 hrs for the 2011 growing season (W126), and mean plant injury values (INJP) for out-planted Tree-of-Heaven seedlings from the OR seed source per sampling date in survey. Error bars represent ± SE. Northeastern Naturalist Vol. 21, No. 4 L.K. Seiler, D.R. Decoteau, and D.D. Davis 2014 549 There were also periods of lower ozone-concentration events from 1–17 August and 3–13 September. However, ozone-induced foliar injury increased in an accumulative pattern even during periods of (relatively) low ambient ozone concentration. Ozone-induced foliar injury was observed on most Tree-of-Heaven seedlings when the SUM60 reached approximately 14,000 ppb.hrs on about 8 August. SUM60 at the conclusion of the survey was 15,232 ppb.hrs. The cumulative sum of weighted hourly concentrations from 800–2000 hr daily (W126 index) increased more steadily over the course of the growing season than the SUM60 index (Fig. 3). Patterns of high- and low-concentration ozone events indicated by the SUM60 index also occurred in the W126 index values. The final W126 ozone index value for central PA in 2011 was 11,600 ppb.hrs. Discussion Our results suggest a range in ozone susceptibility among Tree-of-Heaven populations. Tree-of-Heaven seedlings from the OR seed source were significantly more sensitive to ozone than seedlings from the other seed sources (Table 1, Fig. 1). This finding might have been related to unknown environmental factors that govern ozone sensitivity (Bennett et al. 2006, Hildebrand et al. 1996), but also might have been related to the evolution of tolerance, or lack thereof, to ambient ozone among the various plant populations (Berrang et al. 1989, 1991). However, long-term ambient ozone-monitoring data is not available to test the latter hypothesis. The OR seedlings were smaller compared to seedlings from other seed sources. Ozone susceptibility in plants is governed by a complex set of factors, ranging from stomatal uptake/flux at the whole-plant level (Orendovici et al. 2007) to activity/efficiency of antioxidants at the cell-membrane level (Krupa 1997, Krupa et al. 2001). Small plants, such as the OR seedlings, may have fewer or less efficient antioxidants, making them less able to repair ozone injury as compared to larger, healthier plants (Krupa et al. 2001). Similarly, Tree-of-Heaven seedlings were smaller (mean initial height = 12.1 cm) at the start of the 2010 ozone exposure, than those in the 2011 exposure experiment (mean initial height = 23.4 cm). After 3 weeks of exposure to 120 ppb ozone in 2010, the initially smaller Tree-of-Heaven seedlings had a mean INJP of 2705 compared to a mean of 627 for the larger seedlings used in 2011. This result might be related to plant size, but it is confounded because the 2010 exposure also occurred later in the growing season when plants are generally more sensitive to ozone. In addition, differences in unknown environmental factors between 2010 and 2011 may have influenced the different results from the 2 years. Tree-of-Heaven seedlings exposed to ozone in the greenhouse chambers displayed greater foliar injury with increasing ozone concentration, except seedlings exposed to 40 ppb ozone, which did not differ from the control. The regression model from the greenhouse studies indicated a positive linear relationship between foliar injury and ozone concentration, suggesting that ozone-induced foliar injury is indicative of the relative concentration of ozone in the area, which suggests that a bioindicator may be useful as a biomonitor for ozone damage (Manning and Federer 1980). Northeastern Naturalist 550 L.K. Seiler, D.R. Decoteau, and D.D. Davis 2014 Vol. 21, No. 4 Foliar ozone injury exhibited a significantly negative linear relationship with exposure temperature. Davis and Wood (1973) reported a similar strong negative linear relationship between exposure temperature and ozone-induced plant injury. However, both studies were conduced in chambers utilizing watered potted seedlings, which may not reflect the complicated environmental relationships found in nature. That is, ozone enters the plant during gas uptake through leaf stomata (USEPA 2006), which usually close during times of moisture stress (Reynolds- Henne 2010). In the field, moisture stress is induced by low rainfall and high ambient temperatures, resulting in stomatal closure and reduced ozone uptake (Grulke et al. 2003, Matyssek et al. 2006, Panek 2004), even in the presence of high ambient ozone levels (Davis and Oredovici 2006, Showman 1991, Yuska et al. 2003). Therefore, the regression equation derived from chamber data may have limited utility in predicting outcomes for field-grown plants. In the field, Tree-of-Heaven seedlings displayed ozone-induced black stipple early in the monitoring period, but displayed a tan stipple later in the survey period. In contrast, seedlings exposed to ozone within CSTR chambers under greenhouse conditions displayed only tan stipple. The two ozone-induced responses have been observed previously on Tree-of-Heaven in both the laboratory (Davis and Coppolino 1974) and field setting (Davis and Orendovici 2006). Treeof- Heaven seedlings, saplings, and trees exposed to ambient ozone in the field most commonly exhibit the classic black stipple, which may occasionally change to tan by late summer (D.D. Davis, pers. observ.). In contrast, the potted Tree-of- Heaven seedlings exposed to ozone in CSTR chambers in this study displayed only tan stipple. The mechanism for induction of two different symptom types on Tree-of-Heaven is unknown, but may be influenced by ozone dose, leaf maturity, or environmental factors, such as light intensity. Whatever the reason, when using Tree-of-Heaven as an ozone bioindicator, field evaluations must recognize that ozone can induce both symptom types. The EPA current secondary standard for ozone pollution, which was developed to protect crops and vegetation against injury, states that the 3-year mean of the 4th-highest daily maximum 8-hr average ozone concentrations measured at each monitor within an area should not exceed 75 ppb. Ambient ozone was monitored at two locations near our field-study site. The ozone monitor located ~1.1 km from our study site revealed a 3-year mean of the 4th-highest daily maximum 8-hr-average ozone concentrations of 71.5 ppb for 2009–2011, which is less than EPA’s NAAQS for ozone pollution. This result suggests that the current NAAQS for ozone may not be sufficient to protect sensitive vegetation from ozone-induced injury in central PA. However, ozone levels were relatively high in 2011, and the monitor yielded a 1-year 4th-highest daily maximum 8-hr average ozone concentration of 76 ppb. If ozone concentrations remain at this level or increase, our study region will be in violation of the NAAQS for ozone. For comparison, data from EPA AIRS Site #42- 027-0100, located 16.8 km northeast of our study site, revealed a 1-year 4th-highest daily maximum 8-hr average ozone concentration of 73 ppb for 2011, which is less than the NAAQS for ozone. Also, it should be noted that potted Tree-of-Heaven Northeastern Naturalist Vol. 21, No. 4 L.K. Seiler, D.R. Decoteau, and D.D. Davis 2014 551 seedlings in the greenhouse developed symptoms following exposure to 60 ppb and 75 ppb ozone, indicating that the species is inherently sensitive to ozone. We did not observe ozone-induced foliar injury on most Tree-of-Heaven plants in the field until the ambient SUM60 ozone values reached approximately 14,000 ppb.hrs. This finding implies that a SUM60 value of 14,000 ppb.hrs may be near the threshold at which Tree-of-Heaven seedlings initiate foliar symptoms, at least under the environmental conditions of central PA present in 2011 and considering the fact that the seedlings were watered occasionally and thus were probably not moisture-stressed. This threshold is slightly greater than other estimates. For example, SUM60 ozone needed to cause foliar symptoms on other species of ozone-sensitive plants was estimated at approximately 9000 ppb.hrs or less than 10,000 ppb.hrs in Missouri and South Carolina, respectively (Davis 2009, Davis 2011). However, we also observed that ozone-induced foliar injury increased even in periods of low ambient ozone. Therefore, there may be lag time between an increase in SUM60 ozone and development of ozone injury in sensitive plants, possibly influenced by environmental conditions such as soil moisture (Davis and Oredovici 2006, Showman 1991, Yuska et al. 2003). Ozone injury was more severe as the growing season progressed, which is a result supported by earlier research in our lab (Davis and Coppolino 1974). However, this observation may not always be related to ozone concentration, because older leaves are often more affected by ozone than younger leaves (Skelly 2000). Thus, the increasing sensitivity to ozone over time, as observed in our survey, may be related to the interaction of leaf age and ozone concentration. Alternatively, as autumn approaches, deciduous trees begin to shut down, shed their leaves, and terminate physiological functions; a process that likely makes the leaves more vulnerable to ozone injury. Further research is needed to identify the important factors that determine the incidence and severity of ozone-induced foliar damage, such as mean, peak, or cumulative ozone concentration; exposure duration; species sensitivity; plant or leaf maturity; and various environmental factors. Results of this study contribute to our knowledge of exposure/response relationships between ozone levels and foliar symptoms on ozone-sensitive plant species, and provide data that can inform the process of setting NAAQS in the US for photochemical oxidants and ozone (USEPA 2006, 2007, 2010). The EPA has considered an ozone W126 value secondary standard of 7000 to 15,000 ppb.hrs (USEPA 2007, 2010). The 2011 W126 value for our study area was 11,600 ppb.hrs. According to our field observations, a secondary standard of approximately 11,000 ppb.hrs may not be sufficient to protect sensitive vegetation from ozone-induced foliar damage under moist conditions. Therefore, if a W126 secondary standard is adopted, a value near or lower than 10,000 ppb.hrs might be best. However, Vander Heyden et al. (2000) suggested that a value of 10,000 ppb.hrs might be too high to protect ozone-sensitive vegetation. Similarly, Gravano et al. (2003) reported that ozone injury on a watered Tree-of-Heaven clone was initiated at only 4000 to 5000 ppb.hrs in central Italy. Clearly, additional field research is needed to accurately determine a realistic NAAQS for photochemical oxidants and ozone to protect ozone-sensitive plant species in the United States. Northeastern Naturalist 552 L.K. Seiler, D.R. Decoteau, and D.D. Davis 2014 Vol. 21, No. 4 Acknowledgments We thank M. Kaye, J. Savage, J. Ferdinand, A. Hoffman, K. Webber, C. Fitzpatrick, and J. Millican for their help. We are also grateful for funding from the Pennsylvania Department of Environmental Protection, Bureau of Air Quality, Harrisburg, PA. Literature Cited Bennett, J.P., E.A. Jepsen, and J.A. Roth. 2006. Field responses of Prunus serotina and Asclepias syriaca to ozone around southern Lake Michigan. Environmental Pollution 142:354–366. Berrang, P., D.F. Karnosky, and J.P. Bennett. 1989. Natural selection for ozone tolerance in Populus tremuloides: Field verification. Canadian Journal of Forestry Research 19:519–522. Berrang, P., D.F. Karnosky, and J.P. Bennett. 1991. Natural selection for ozone tolerance in Populus tremuloides: An evaluation of nationwide trends. Canadian Journal of Forestry Research 21:1091–1097. Davis, D.D. 2007a. Ozone-induced symptoms on vegetation within the Moosehorn National Wildlife Refuge in Maine. Northeastern Naturalist 14:403–414. Davis, D.D. 2007b. Ozone injury to vegetation within the Seney National Wildlife Refuge in the Upper Peninsula of Michigan. Northeastern Naturalist 14:415–424. Davis, D.D. 2009. Ozone-induced stipple on plants in the Cape Romain National Wildlife Refuge, South Carolina. Southeastern Naturalist 8:471–478. Davis, D.D. 2011. Ozone-induced leaf symptoms on vegetation in the Mingo National Wildlife Refuge Missouri. Northeastern Naturalist 18:115–122. Davis, D.D., and J.B. Coppolino. 1974. Relative ozone susceptibility of selected woody ornamentals. HortScience 9:537–539. Davis, D.D., and T. Orendovici. 2006. Incidence of ozone symptoms on vegetation within a national wildlife refuge in New Jersey, USA. Environmental Pollution 143:555–564. Davis, D.D., and F.A. Wood. 1973. The influence of environmental factors on the sensitivity of Virginia Pine to ozone. Phytopathology 63:371–376. Gravano, E., V. Giulietti, R. Desotgiu, F. Bussotti, P. Grossoni, G. Gerosa, and C. Tani. 2003. Foliar response of an Ailanthus altissima clone in two sites with different levels of ozone-pollution. Environmental Pollution 121:137–146. Grulke, N.E., R. Johnson, A. Esperanza, D. Jones, T. Nguyen, S. Posch, and M. Tausz. 2003. Canopy transpiration of Jeffrey Pine in mesic and xeric microsites: O3 uptake and injury response. Trees-Structure and Function 17:292–298. Heck, W.W., R.B. Philbeck, and J.A. Dunning. 1975. A continuous stirred tank reactor (CSTR) system for exposing plants to gaseous contaminants: Principles, specifications, construction and operation. USDA-ARS Pub. No. ARS-S-181, New Orleans, LA. Hildebrand, E., J.M. Skelly, and T.S. Fredericksen. 1996. Foliar response of ozone-sensitive hardwood-tree species from 1991 to 1993 in the Shenandoah National Park, VA. Canadian Journal of Forestry Research 26:658–669. Horsfall, J.G., and R.W. Barratt. 1945. An improved grading system for measuring plant disease. Phytopathology 35:655. Kasson, M.T., M.D. Davis, and D.D. Davis. 2013. The invasive Ailanthus altissima in Pennsylvania: History, distribution, and growth patterns. Northeastern Naturalist 20 (Monograph No. 10):1–60. Krupa, S.V. 1997. Air Pollution, People, and Plants. The American Phytopathological Society, St. Paul, MN, 197 pp. Northeastern Naturalist Vol. 21, No. 4 L.K. Seiler, D.R. Decoteau, and D.D. Davis 2014 553 Krupa, S.V., M.T. McGrath, C.P. Andersen, F.L. Booker, K.O. Burkey, A.H. Chappelka, B.I. Chevone, E.J. Pell, and B.A. Zilinskas. 2001. Ambient ozone and plant health. The American Phytopathological Society 85:4–12. Manning, W.J., and W.A. Federer. 1980. Biomonitoring Air Pollutants with Plants. Applied Science Publishers, UK. 142 pp. Matyssek, R., D. LeThiec, M. Low, P. Dizengremel, A.J. Nunn, and K.H. Haberle. 2006. Interactions between drought and O3 stress in forest trees. Plant Biology 1:11–17. Orendovici, T., J.M. Skelly, D.D. Davis, J.A. Ferdinand, J.E. Savage, and R.E. Stevenson. 2007. Ozone uptake (flux) as it relates to ozone-induced foliar symptoms of Prunus serotina and Populus maximowizii x trichocarpa. Environmental Pollution 151:79–92. Orendovici-Best, T., J.M. Skelly, and D.D. Davis. 2010. Spatial and temporal patterns of ground-level ozone within north-central Pennsylvania forests. Northeastern Naturalist 17:247–260. Panek, J.A. 2004. Ozone uptake, water loss and carbon-exchange dynamics in annually drought-stressed Pinus ponderosa forests: Measured trends and parameters for uptake modeling. Tree Physiology 24:277–290. Reynolds-Henne, C.E., A. Langenegger, J. Mani, N. Schenk, A. Zumsteg, and U. Feller. 2010. Interactions between temperature, drought, and stomatal opening in legumes. Environmental and Experimental Botany 68:37–43. Richards, B.L., J.T. Middleton, and W.B. Hewitt. 1958. Air pollution with relation to agronomic crops. V. Oxidant stipple of grape. Agronomy Journal 50:559–561. SAS Institute. 2011. SAS software, Version 9.1 of the SAS System for Unix. Cary, NC. Showman, R.E. 1991. A comparison of ozone injury to vegetation during moist and drought years. Journal of Air and Waste Management Association 41:63–64. Skelly, J.M. 2000. Tropospheric ozone and its importance to forest and natural plant communities of the northeastern United States. Northeastern Naturalist 7:221–236. Skelly, J.M., D.D. Davis, W. Merrill, E.A. Cameron, H.D. Brown, D.B. Drummond, and L.S. Dochinger. 1987. Diagnosing injury to eastern forest trees. USDA Forest Service, Forest Pest Management, Atlanta, GA, and The Pennsylvania State University, College of Agricultural Sciences, Department of Plant Pathology, University Park, PA, 22 pp. United States Department of the Interior (USDOI). 2003. Ozone-sensitive plant species on National Park Service and US Fish and Wildlife Service lands: Results of a 24–25 June 2003 workshop, Baltimore, MD. NPS/NRARD/NRR-2003/0, Denver, CO, 21 pp. United States Environmental Protection Agency (USEPA). 2003. Ozone: Good up high, bad nearby. EPA-451/K-03-001, Washington, DC. USEPA. 2006. Air quality criteria for ozone and related photochemical oxidants, volume III of III. EPA 600/R-05/004cF, Research Triangle Park, NC. USEPA. 2007. Review of the national ambient air quality standards for ozone: Policy assessment of scientific and technical information. EPA-452/R-07-007, Research Triangle Park, NC. USEPA. 2010. Proposed rule, national ambient air quality standards for ozone. Federal Register/Vol. 75, No. 11/Tuesday, 19 January 2010, Research Triangle Park, NC. Yuska, D.E., J.M. Skelly, J.A. Ferdinand, R.E. Stevenson, J.E. Savage, J.D. Mulik, and A. Hines. 2003. Use of bioindicators and passive sampling devices to evaluate ambient ozone concentrations in north central Pennsylvania. Environmental Pollution 125:71–80.