Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Tropospheric ozone in the Lower Fraser Valley, British Columbia and the threat of injury to forest plants Krzyzanowski, Judi 2003

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2003-0345.pdf [ 37.13MB ]
Metadata
JSON: 831-1.0091113.json
JSON-LD: 831-1.0091113-ld.json
RDF/XML (Pretty): 831-1.0091113-rdf.xml
RDF/JSON: 831-1.0091113-rdf.json
Turtle: 831-1.0091113-turtle.txt
N-Triples: 831-1.0091113-rdf-ntriples.txt
Original Record: 831-1.0091113-source.json
Full Text
831-1.0091113-fulltext.txt
Citation
831-1.0091113.ris

Full Text

T R O P O S P H E R I C O Z O N E I N T H E L O W E R FRASER V A L L E Y , B R I T I S H C O L U M B I A A N D T H E T H R E A T O F IN JURY T O F O R E S T P L A N T S by J U D I K R Z Y Z A N O W S K I B.Sc, York University, 2001 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F S C I E N C E in T H E F A C U L T Y O F G R A D U A T E STUDIES (The Department of Geography) W e accept this thesis as confonning to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A August 2003 © Judi Krzyzanowski, 2003 I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r an advanced degree a t t h e U n i v e r s i t y o f B r i t i s h C o l u m b i a , I agree t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d by the head of my department o r by h i s o r h e r r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l n o t be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . The U n i v e r s i t y o f B r i t i s h C olumbia Vancouver, Canada Department A B S T R A C T During the summers of 2001 and 2002 ambient ozone levels were measured as hourly averages in parts per bill ion (ppb) at four sites of differing elevation (200, 400, 600 and 1200 m) in the Lower Fraser Valley (LFV), British Columbia. A l l sites were located in forest clearings and experienced hourly averages as low as Oppb, and >70ppb. Mean seasonal concentrations show an increase in ambient ozone with elevation due to consistently high nocturnal concentrations and lack of diurnal variation at higher altitudes. Diurnal patterns are in agreement with previous studies showing a peak in concentration in the late afternoon, and a morning increase due to photochemical production and residual layer down mixing. The occurrence of an upper level ridge coinciding with a thermal trough along the coast, cause above average ozone levels to occur, and may cause the National Ambient A i r Quality Objective of 82ppb to be exceeded. Cumulative ozone exposures were measured with height in a forest canopy using O G A W A passive samplers mounted to a 10.5m tower. A strong power-law increase in ozone with height was found due to a number of potential processes including deposition and chemical destruction at the surface, uptake by vegetation, and dynamic stability inhibiting down mixing. This relationship shows plants in the L F V may be acting as an effective ozone sink. Plants uptake ozone direcdy through leaf stomates where the pollutant may direcdy injure foliage. A preliminary survey of native shrubs exhibiting visible ozone injury symptoms suggests that current concentrations of tropospheric ozone in the L F V may be high enough to cause injury to forest species, however more research is required in order to determine the threat of ozone to these economically and culturally important forests. i i T A B L E O F C O N T E N T S Page Title Page i Abstract i i Table of Contents iii List of Figures v List of Tables vi Acknowledgements vii 1.0 Introduction 1 2.0 literature Review 3 2.1 A i r Quality in the Lower Fraser Valley , B C 4 2.2 Ozone Chemistry and Formation 9 2.3 Ozone Variation with Elevation 11 2.4 Ozone Impacts on Forests 14 3.0 Methods 18 3.1 Study Area and Site Characteristics 18 3.2 Continuous Monitoring for Altitudinal Variation 22 3.3 Passive Monitoring for With in Canopy Variation 25 3.4 Characterisation of Ozone Indices 31 3.5 Vegetation Sampling for Foliar Injury 32 4.0 Altitudinal and Seasonal Variation in Ozone 36 4.1 Altitudinal Comparison 36 4.2 Diurnal Trends 40 4.3 Episodic Increases and Synoptic Influence 43 4.3.2 Episode August 8-17, 2001 44 4.3.3 Episode July 19-27, 2002 51 4.4 Summary of Altitudinal and Seasonal Variation 58 i i i 5.0 With in Canopy Variation in Ozone 63 5.1 Passive Sampling Results 63 5.2 Meteorological Results 66 5.3 Factors Affecting within Canopy Ozone Profiles 70 6.0 Impacts to Forest Vegetation 75 6.1 Seasonal Development of Cumulative Ozone 75 6.1.1 AOT40 : 76 6.1.2 SUM06 78 6.1.1 Discussion of Indices 81 6.2 Foliar Symptoms and Injury Development 82 6.3 Discussion of Injury to Vegetation 91 6.4 Summary of Impacts to Forest Vegetation 95 7.0 Conclusions 98 7.1 Summary of Findings 98 7.2 Future Research 100 7.3 Applications and Recommendations 103 Appendix A : Laboratory Protocol 105 Appendix B: Hourly Ozone Data 115 B. l Summer of 2001 T ime Series , 115 B. 2 Summer of 2002 T ime Series 120 Appendix C: Injury to Vegetation 125 C. l Photos and Horsfall-Barratt Injury Scores 125 C.2 Injury Symptoms and Severity 146 Works G ted 149 iv L IST O F F IGURES Page 2.1 Map of the Georgia Basin with study area location 5 2.2 Localised flow patterns characteristic of urban complex coastal terrain 8 3.1 Map of study area in relation to the L F V showing field site locations 19 3.2 Site set-up of continuous monitoring locations ; 24 3.3 a) Photograph of 10.5m tower in the forest canopy 27 b) Diagram of tower including location of samplers and other devices 28 3.4 Photos of Ogawa passive ozone samplers and components 29 3.5 Typical foliar injury symptoms on Comus sanguinea 33 4.1 Box and whisker plot of 2001 altitudinal variation in ozone concentration 37 4.2 Box and whisker plot of 2002 altitudinal variation in ozone concentration 39 4.3 Mean seasonal hourly average ozone concentrations for 2001 41 4.4 Mean seasonal hourly average ozone concentrations for 2002 42 4.5 Hourly average ozone concentrations for the August 8-17, 2001 episode 45 4.6 Regional hourly average ozone concentrations for the August 8-17, 2001 episode 47 4.7 One day means of sea level pressure for August 8*, 12*, and 17*, 2001 49 4.8 One day means of 500mb geopotential height for August 8*, 12*, and 17*, 2002 50 4.9 Hourly average ozone concentrations for the July 19* - 28*, 2002 episode 52 4.10 Regional hourly average ozone concentrations for the July 19* - 28*, 2002 episode .. 54 4.11 One day means of sea level pressure for July 19*, 23"*, and 28*, 2002 56 4.12 One day means of 500mb geopotential height for July 19*, 23 r d, and 28*, 2002 57 5.1 . Power-law profiles of hourly average ozone concentration in a forest canopy 64 5.2 With in canopy profiles of biweekly mean temperature and relative humidity 66 5.3 Seasonal trends in temperature and relative humidity 68 5.4 Seasonal trends in windspeed and wind direction 69 5.5 Schematic diagram of factors influencing ozone distribution in a forest canopy 73 6.1 Cumulative A O T 4 0 at all sites for the 2001 study season 77 6.2 Cumulative A O T 4 0 at all sires for the 2002 study season 78 6.3 Cumulative SUM06 at all sites for the 2001 study season 79 6.4 Cumulative SUM06 at all sites for the 2002 study season 80 6.5 Plots of H B score progression recorded for each of the monitored plants 85 v L I ST O F T A B L E S Page 3.1 Species chosen for the monitoring of visible ozone injury symptoms 35 4.1 Summary of ozone concentrations for 2001 38 4.2 Summary of ozone concentrations for 2002 40 6.1 Description of initial foliar injury symptoms for each of the monitored plants 83 vi A C K N O W L E D G E M E N T S I would like to share my gratitude with the numerous individuals and organisations who made this work possible. Firstly, I would like to thank my academic committee for their contributions to this research. The utmost thanks to my supervisor Ian McKendry whose support and meteorological wisdom have allowed me to succeed in this endeavour - I thank Ian for his faith in my sanity and endless encouragement I give special thanks to John Innes for his inspirational guidance and his extensive insight into ozone injury. Also, thanks to Douw Steyn whose tough questions reminded me why I am doing this in the first place. I would like to thank N S E R C for providing funding to Ian McKendry for this research, and to the Department of Geography at the U B C for providing my personal funding as a research assistantship throughout the field season. Thanks to the CFS (Canadian Forest Service) for providing financing, equipment for continuous monitoring, and for entrusting their project in my hands. Thanks to Nick Humphreys and Leo Unger of the CFS for teaching me everything they know. Many thanks to Mike Brauer for allowing me (the cursed one) to use his lab, passive samplers and ion chromatographer. Thanks to the B C Ministry of Water, Land and A i r Protection for providing regional ozone and meteorological data. Thanks to Campbell Scientific for their technical support. I offer my utmost appreciation to those who aided me in the field. Emily Fischer who without I would still be lost in the woods, hungry and collecting data; Craig Clements whose presence always shortened the climb; and both Cindy Walsh and Phoebe Jackson for helping to erect the tower. Thanks to both Adrian Kostic and Matt Schnurr for giving up their days to help ensure the success of this project. v i i Thanks to all of my colleagues in the Geography Department (et al.) for both understanding my pain and remembering how to act like teenagers - especially when on a picket line. I thank all the support staff and faculty of U B C Geography for helping make these two years a surprisingly pleasant experience. I would also like to thank my mother Jane and sister Nancy for believing in me, both academically and otherwise. Lasdy, to those who have in some way aided this research but have not been mentioned, from the extended meteorological community to my cat Luna, I am truly grateful. v i i i 1.0 I N T R O D U C T I O N Photochemical smog and its principle component, ozone, are widespread atmospheric pollutants. The elevated concentrations of tropospheric ozone in 'smog' are known to cause direct injury to vegetation. Since background and periodic concentrations of tropospheric ozone are expected to increase further over the next few decades, this study attempts to ascertain whether forests of the Lower Fraser Valley may be at risk of damage due to ozone. There has been some focus on the distribution of tropospheric ozone with elevation, and ozone levels within and around the LFV . However, this study documents for the first time, altitudinal distribution of Os along south facing slopes in the LFV , and identifies potential processes involved in ozone transport in the region. Previous studies have shown ozone levels to be particularly high within the range of intermediate altitudes (400 - 1800m) (Miller and Arbaugh 2000; Sandroni et al. 1994). The monitoring of summertime ozone at sites of varying altitude (200 - 1200m) is used here to establish trends in ozone distribution on slopes at the mesoscale. This is also a preliminary study of how ambient ozone levels vary vertically within forest canopies of the LFV , at the microscale. Finally, this study attempts to understand how ambient ozone concentrations measured at both scales may influence the exposure of natural forest vegetation to this pollutant This research has four objectives: 1) T o determine concentrations of ambient ozone in the L F V and to document diurnal, seasonal and altitudinal patterns, by monitoring hourly-mean ozone concentrations on the south-facing slopes of the LFV . 2) T o construct a characteristic summertime, vertical ozone profile within a forest canopy by measuring cumulative ozone concentrations at various canopy heights. 1 3) T o document any potential' ozone injury to native vegetation incurred throughout the growing season by visually examining forest plants for typical injury symptoms, and subsequently monitoring individuals for symptom development 4) T o determine whether the forests on the south-facing slopes of the L F V are at risk of injury or alteration due to current ambient ozone concentrations through the use of air quality standards, and native forest plants as bioindicators. Ozone levels were monitored during summer months, from June to October 2001 and 2002, when tropospheric ozone concentrations are high due to both long periods of sunlight, and the high-pressure systems dominating the region at this time of year. The summer months correspond also, to the vegetative growing season when gaseous uptake by plants is the highest, This represents the most opportune time to monitor ozone with respect to forests - the resilience of which is a crucial part of our economy and environment The following chapter, chapter 2.0, is a review of literature pertaining to ozone and forest health, including air quality in the LFV; ozone chemistry and formation; ozone variation with elevation; and ozone impacts on forest health from around the world. In subsequent chapters the results of the 2001 and 2002 field studies will be discussed. Chapter 3.0 will describe the methodology followed to carry out the four objectives stated above. Chapters 4.0, 5.0, 6.0 and 7.0 will be the results and discussion obtained from the examination of objectives 1, 2, 3 and 4, respectively. Lastly, chapter 8.0 will conclude with a summary of findings, a qualitative correlation of the four objectives and recommendation for future air-quality standards and forest management practices. 2 2.0 L I T E R A T U R E R E V I E W Tropospheric ozone, the principle component of photochemical smog, is a globally common atmospheric pollutant. There are both natural and anthropogenic sources of ground levels ozone. Natural sources include hghtning and stratospheric intrusion. Also, volatile hydrocarbons and oxides of nitrogen - key ingredients in ozone formation at the earth's surface - are produced by biological processes. Despite natural sources that contribute to a global background concentration between 20-40ppb, human activity is the predominant contributor to elevated ozone in the lower troposphere (Krupa and Kickert 1997). Elevated concentrations of ground-level ozone are known to cause damage to plants, animals, people and built structures. Potentially injurious levels of ozone in the free troposphere have been documented by studies world-wide. These studies include, but are not limited to, those conducted in northern (Browell et ai. 1994); western (McKendry 1994; McKendry et al. 1997; McKendry et al. 1998); and eastern (McLaughlin 1998) Canada; parts of the United States flacob et al. 1993; Aneja et al. 1994; Kleinman et al. 1994; Cooper and Peterson 2000); Mexico (Fast and Zhong 1998), Europe (Broder et al. 1981; Wanner et al. 1993; K lemm et al. 1998); and globally via satellite (Fishman et al. 1990). Photochemical smog was first described in Los Angeles in 1943 (Cadle 1971) and ozone levels have increased significandy around the globe over the past half century (Fishman et al. 1990; Puxbaum et al. 1991; McLaughlin, D., 1998). This study is concerned with anthropogenically induced tropospheric ozone in the L F V and whether or not the concentrations are high enough to cause plant injury. The following is a review of critical literature in this field. 3 2.1 A i r Quality in the Lower Fraser Valley, B C The Lower Fraser Valley (LFV) in British Columbia, Canada is a coastal region of complex terrain, creating a unique boundary layer environment into which pollutants are emitted. Ozone is the oxidant of highest concentration in the L F V (Pisano et ai. 1997), mainly from the use of automobiles in the city of Vancouver and the surrounding municipalities (McKendry 1994). Although the urban area of Vancouver, where most of the precursors of photochemical smog are produced has relatively clean air, the pollution is transported eastward to rural areas where ecological damage may be occurring. Figure 2.1 shows the Lower Fraser Valley (LFV) and the surrounding area of the Georgia Basin. The study area is represented by the red-cross. The study area is east or downwind of the urban area Vancouver. Summer-time ozone levels are usually higher at downwind rural locations, than at their urban source (Brace and Peterson 1998; Cooper and Peterson 2000). As distance from the source increases pollutant layering represents larger scale atmospheric stratification (Reiter 1991; McKendry and Lundgren 2000). Seasonally, ozone levels are the highest during spring and summer months due to clear skies, high temperatures and increased solar radiation (McKendry et al. 1997; Sillman et al. 1990) favouring the photodissociation of nitrogen dioxide. Ozone levels begin to increase in the free troposphere in Apr i l due to the accumulation of ozone precursors (oxides of nitrogen) during the winter (Stockwell et al. 1997) and due to the increase in solar radiation at this time (McKendry and Lundgren 2000). Diurnally, tropospheric ozone concentrations typically peak in the afternoon (the time of most intense solar heating) and have a minimum at night in the stable nocturnal boundary layer (Pisano et al. 1997). 4 Figure 2.1- Map of the Georgia Basin. The red cross marks the general location of the study area. In the LFV , the vertical distribution of ozone in the lower troposphere is determined by boundary layer stability and localised wind patterns. A diurnally cycling complex layering of pollutants results. Processes leading to the development of pollutant layers in the region are illustrated by Figure 2.2. In addition to localised processes that are unique to areas of complex coastal terrain, vertical ozone distribution in the lower troposphere also reflects larger-scale boundary layer stratification to be further discussed in section 2.3. Noctumally, elevated pollutant layers become trapped either within, or above, elevated surface inversions (Stall 2000; McKendry et al. 1997). A t night in the L F V the surface atmosphere is characteristically a stable nocturnal boundary layer (NBL) with a temperature inversion above leading to increased stability, and dry deposition at the surface removing 5 pollutants (McKendry and Lundgren 2000). There are some processes leading to pollutant layering that are unique to areas of urbanised complex coastal terrain, such as the LFV . The following discussion refers to all pollutants and it is important to note that when mentioning pollutant plumes near urban areas ozone precursors are present in large concentrations, but ozone is not formed until the plume has moved further downwind. Figure 2.2 is a schematic diagram showing processes leading to complex pollutant layering in regions of urban complex coastal terrain, from McKendry and Lundgren (2000). © Represents the offshore advection of pollutants (including NOa and other Oa precursors) from the city in the early morning. This flow regime is a remnant of the nocturnal katabatic winds characteristic of complex terrain and the land breeze characteristic of coastal regions. The layer shown in ® is kept isolated from the surface via the cold stable surface air and kept low by the inversion layer aloft As the sun rises towards solar noon and surface heating occurs, the flow reverses and an onshore sea breeze replaces the offshore land breeze. The sea breeze encounters the urban plume and pushes it up wards in a process described as sea breeze lofting ©. Offshore subsidence may allow these pollutants to be recirculated within the sea breeze regime. Likewise, as heating begins an up slope anabatic wind replaces the down slope katabatic flow and pollutants such as ozone upon reaching the slope, are blown upwards and may be injected above the inversion layer by this mountain venting CD. In the afternoon the sea breeze front extends downwind of the city and undercuts the polluted layer with clean, dense, cool air ©. Pollutants exist aloft, above the city, isolated from the free troposphere via subsidence. Just downwind of the urban area, new pollutants are emitted near the surface and convection and anabatic winds continue to induce the up slope flow of atmospheric pollutants. The slope return flow that exists in a daytime mountain 6 circulation cell injects some pollution/ozone into the stable inversion layer ©, while mountain venting continues © and succeeds in allowing pollutants to penetrate the inversion and enter the free troposphere. As evening stabilisation occurs, the presence of flow-blocking mountains may keep the classic stable layer and associated residual layer, from developing. Instead, when mountains are present the potential energy from the daytime mixed layer, coupled with cooling and subsidence aloft, leads to a horizontal injection of pollutants into the inversion layer ®. This creates elevated layers of pollutants in the evening in areas of complex terrain (McKendry and Lundgren 2000). Currendy, the National Ambient Hourly Os A i r Quality Objective is 82 ppb (Stull 2000; McKendry 1994) which was determined jointly by various levels of Canadian government This is considered the maximum hourly average concentration of ozone that can occur before there are severe effects on the health of humans and the environment. This maximum is exceeded in the L F V approximately eight days each summer at one or more of the 24 monitoring stations of the L F V (Pryor et al. 1995). This exceedance is due to synoptic scale high-pressure systems (McKendry and Lundgren 2000; McKendry 1994) that adversely effect local air quality. 7 <ZZWind Figure 2.2 - Some localised How patterns characteristic of urbanised complex coastal terrain (the LFV), that, along with typical boundary conditions, lead to persistent layers of elevated pollutants. Each numbered process is described within the text. Source: McKendry and Lundgren 2000 8 2.2 Ozone Chemistry and Formation Tropospheric ozone is a secondary pollutant formed by complex reactions between primary pollutants - produced chiefly by fossil fuel combustion. The principal reaction responsible for ozone formation in the troposphere is the photolysis of nitrogen dioxide by ultraviolet solar radiation. This results in nitrogen oxide and a free oxygen atom, which reacts with molecular oxygen in the atmosphere to produce ozone (Glisten et al. 1998; Pisano et al. 1997; Neu etal. 1994; Comrie etal. 1990). The production of ozone from NO2 photodissociation is illustrated by the following chemical reactions: NO2 + /1V -* N O +0 (Rl) O + O2 + M -» O 3 + M (R2) 0 3 + N O -> NO* + O* (R3) where hv represents ultraviolet radiation and M is any molecule (such as nitrogen or oxygen) that helps to stabilise the ozone molecule by dissipating the energy released through its creation (Innes etal. 2001; Pisano etal. 1997; Lefohn 1992; Comrie etal. 1990). If reactions 1 through 3 repeat, there is no net accumulation of ozone, thus representing a photostationary state (Arya 1999). This process is known as "nitrogen scavenging". However, in the presence of volatile organic compounds (VOCs) the photostationary state of equilibrium breaks down and nitrogen scavenging does not occur. The emission of VOCs , most importandy anthropogenic hydrocarbons, is primarily associated with the incomplete combustion of fossil fuels by mobile sources. T o a much lesser extent in the L F V industrial activity, and natural sources such as isoprene and turpenes naturally emitted from forests may also contribute to V O C emission . In the presence of V O C s , N O effectively oxidized to NO2 without the dissociation of ozone. In 9 this case equation 3) does not occur. Instead, more NO2 is produced fuelling Os production, and existing Os molecules remain intact This leads to net accumulation of ozone in the troposphere (Comrie etal. 1990; Neu eta/. 1994; Pisano era/. 1997; Giisten era/. 1998; Arya 1999) illustrated by R4), which is then followed by 1) etc. HCO*, represent oxygenated hydrocarbons with a variable number of oxygen atoms. HCOx + N O -* N X X + H C O , (R4) Ozone production is hmited by the precursor in lowest supply i.e. either NO* or hydrocarbons (Sillman 1993). These reactions produce tropospheric ozone solely in the daytime, since short-wave solar radiation is required for NO2 photodissociation (Rl). Conditions favouring the formation of tropospheric ozone and other components of photochemical smog include light winds, low cloud cover, low relative humidity and high solar insolation (Edmonds and Basabe 1989; Aneja era/. 1994; Giisten etal. 1998). High temperatures associated with the above mentioned fair weather conditions enhance the chemical production of ozone by reactions R l and R2 which are temperature dependent Significant ozone production does not generally occur at temperatures below 15°C (Sillman 1993). Variation in ambient levels of photochemical smog and ozone are strongly linked with the annual cycles of solar radiation, since high intensity short-wave radiation is necessary to facilitate the photolysis of NO2 in R l (Oke 1987). The highest levels of Os in non-urban areas are associated with urban and industrial plumes. Although deposition and chemical destruction reduce Os concentrations near the earth's surface, the lifetime of this pollutant in the lower troposphere may extend from several hours to several days, meaning ozone may be transported great distances (Comrie 1990; 10 Sillman 1993). The lifetime of an ozone molecule depends on the availability of solar radiation, water vapour (Fishman et al. 1991), NOx and V O C s (Sillman 1993). In mid-latitudes, synoptic scale systems such as anti-cyclones and frontal passage, strongly affect the meteorological variables controlling air quality and may cause above average pollutant levels in an 'episode' (McKendry 1994). These meteorological conditions create the warm, dry and sunny environment mentioned above, that favours ozone production. Aneja (1994) defines an ozone episode as a period when hourly averages exceeded 70ppb for more than 8 hours. In the L F V ozone episodes are related to anticyclonic upper-level ridges that produce the stagnation of air over a large area (McKendry et ai. 1998; Comrie 1990). These systems also promote shallow boundary layer development due to subsidence, thereby concentrating air pollutants near the ground for extended periods of time (Oke 1987; Stall 2000). 2.3 Ozone Variation with Elevation The L F V has been documented as having elevated ozone levels throughout the troposphere (McKendry and Lundgren 2000). The presence of mountains with heights exceeding 1500m favours up-slope anabatic winds and reinforces the sea breeze that carries pollutants inland (Millan etal. 1997; Brace and Peterson 1998). Ozone concentrations tend to increase with altitude, low winds, and high temperatures, in the boundary layer during the summer (Aneja et ai. 1994; McKendry et al. 1998; Brace and Peterson 1998; Cooper and Peterson 2000). The vertical transport of ozone witliin the atmosphere is accomplished by both turbulence and local circulation patterns (Lehning et ai. 1998). In this study of south-facing slopes of the LFV , local circulation is predominandy mountain-valley winds, and sea- or land- breezes. 11 During the day, when surface heating promotes mixed layer (ML) development, ozone concentration varies httle with height in the boundary layer (BL). Ozone is deposited near the surface at night within the stable nocturnal boundary layer (NBL). However, above the stable nocturnal boundary layer ozone is isolated by a lack of mixing, from its chemical and depositional surface sinks (McKendry and Lundgren 2000). A t night, ozone concentration is not depleted at high elevations. This ozone persistence with altitude is in part due to the presence of a residual layer (RL) isolated by the stable surface layer below and a capping inversion above (Neu et ai. 1994; McKendry and Lundgren 2000). The residual layer is a remnant of the daytime mixed layer separated from the surface as nocturnal cooling increases the depth of the stable nocturnal boundary layer (NBL). The R L base may be as low as 700m and as high as 1600m depending on the thickness of the N B L . The R L retains the pollutants and the adiabatic lapse rate of the daytime mixed-layer (ML) (Stull 2000). If a high elevation mountain location does not intercept the RL , katabatic winds may advect ozone rich air from aloft down slope (Zaveri et ai. 1995; Brace and Peterson 1998). The R L and other recirculating air masses may contribute between 60% and 100% of daily ozone (Sanz and Mil lan 2000). Factors that may contribute to increases in ozone with elevation include photochemical processes, transport from upwind and local sources, lack of surface removal, and tropospheric mixing or layering. Night-time surface removal (dry deposition and vegetative uptake) does not deplete ozone concentrations at high elevations as it does at lower elevations or on plains as Zaveri et al. (1995) showed with a model forecasting ozone in rural mountainous locations. The nocturnal persistence of ozone is in part due to a continuous supply of ozone from the R L or free troposphere brought to mountains by down-slope flows that counteract any depletion 12 (Zaveri et al. 1995). Low elevation mountain valleys are affected by this down-slope advection to a lesser extent, but experience higher nocturnal ozone concentrations than areas at the same elevation in non-mountainous areas. The model showed valleys to have evident surface removal whereas mountaintops do not (Zaveri et al. 1995). Brace and Peterson (1998), and Cooper and Peterson (2000) found more persistent ozone levels at higher elevation sites in the Cascade Mountains of Washington State. They attributed this lack of depletion to the low concentrations of N O at these locations, inhibiting the nocturnal decay via R3. They also found mountain sites to have a strong positive correlation between elevation and weekly mean ozone exposure. The rate of increase in ozone concentration with elevation was similar throughout the Mount Rainier region (1.3ppbv / 100m elevation gain), however the magnitude of exposure was found to differ spatially (Cooper and Peterson 2000). Both aforementioned studies used passive sampling techniques to measure weekly average ozone concentrations with elevation. Passive samplers measure ozone using a N O coated filter in a sampling apparatus. The N O , when exposed to Os, becomes NOa (see R3). The filter is then analysed to determine the NO/NO2 ratio to give a mean hourly ozone concentration for the sampling period. Both Brace and Peterson (1998), and Peterson and Cooper (2000) used a sampling period of one week's duration. This study uses a two-week long sampling period to measure ozone exposure with height in a forest canopy using passive samplers. Due to the high nocturnal ozone concentrations of mountaintops and valleys, these areas exhibit lesser diurnal variation in ozone than flat, low-elevation sites. The absence of strong diurnal variation of ozone with altitude and the resulting higher mean ozone levels, means that sub-alpine forests are particularly at risk of oxidant damage as they receive a higher 13 daily pollutant dose than forests at lower elevations (Puxbaum et al. 1991; Sandroni et al. 1994; Zaveri et al. 1995; Brace and Peterson 1998). Continuous monitoring of tropospheric ozone in the L F V is necessary to assess both the dose of ozone to forest vegetation, and whether or not the levels are high enough to pose a significant threat to forests. South-facing slopes are particularly susceptible to high ozone levels in the summer due to radiation geometry that effects ozone formation vrithin a plume, and strengthens the anabatic winds that transport pollutants to sub-alpine areas (Sandroni et al. 1994). The forests of sub-alpine areas are particularly at risk of ozone injury. Previous research on the micro-scale variation of ozone concentration with elevation -such as in a forest canopy - though limited, includes studies by Coe et al. (1995); Krupa and Kickert (1997); and Lamaud et al. 2002. However, these studies failed to provide a detailed account of how ozone fluxes or concentrations vary vertically within the forest canopy. This study examines average vertical ozone profiles within a low forest canopy. 2.4 Ozone Impacts on Forests The sub-alpine areas of the L F V are characterised by forested slopes where trees are harvested, thereby representing an economically and culturally important resource to the province of B C . Ozone injury to plants has been well documented since the 1980's, is a problem recognised by a number of national (e.g. Canadian Council of Forest Ministers), and international (e.g. Montreal process, Helsinki process) agreements on sustainable forest management Research on the relationship between air pollution and forest decline has been carried out in much of the United States and Europe. However, with the exception of studies around smelters (e.g. Sudbury, O N and Trail, BC) and this study, the research in this field has 14 been very limited in Canada. Previous Canadian research pertaining to air pollution and forests includes the effects caused by chemicals such as acid rain and heavy metals rather than ozone (McLaughlinetal. 1998). Many fumigation and or open-top chamber experiments have taken place (Brace et al. 1999; Ball et al. 1998; Langebartels et al. 1998; Krupa and Kickert 1997; Samuelson 1994; Pearson and Mansfield 1994; Paakkonen et al. 1993; Amundson et al. 1991; Dobson et al. 1990; Bytnerowicz et al. 1989; Pye 1988; Wang 1986; Hogsett et al. 1985; Reich, and Amundson 1985; Mil ler et al. 1983) usually with potted plants or under laboratory conditions. Controlled field studies include; Beyers et al. 1992; Benoit et al. 1982. Studies of ozone damage that have focussed on natural in situ forest vegetation include Skelly et al. (1987, 1997, 1998, and 1999) and Mil ler (2000). Ozone injury to vegetation can take many forms including; alterations to photosynthetic capacity; pathogen defence mechanisms; biochemical pathways; reproductive viability; and nutrient / biomass allocation (Langebartels et al. 1998; Mc laughl in 1998; Beyers et al. 1992; Lefohn 1992; Sensor et al. 1990; Benoit et al. 1983). Ozone may also lead to an alteration of plant community composition due to differences in resistance / sensitivity in different species (Barbo et al. 1998; Lyons et al. 1997; Bennett and Runeckles 1977). Injury may take the form visible symptoms that can be quantified based on various symptom characteristics. The most common symptoms are inter-veinal necrotic stipple, chlorosis, or a reddening / purpling of the upper leaf surface (Ghosh et al. 1998; Brace et al. 1999; Mil ler and Arbaugh 2000; Sanz and Mil lan 2000; Innes et al. 2001; VanderHeyden et al. 2001). It is important to note that vegetation may incur severe injury due to ozone exposure while displaying no visible symptoms (Krupa and Kickert 1997). Ozone is the only atmospheric pollutant known to cause direct leaf 15 injury and tree mortality (McLaughlin 1998; Comrie 1990) and the uptake of this pollutant occurs right at the leaf cuticle. The diffusion of ozone into plant tissue, and hence ozone uptake by vegetation is dependant on the ozone concentration inside the leaf cells (Krupa and Kickert 1997). Ozone uptake may therefore decrease, as cells become 'ozone saturated'. Peak concentrations are important to note since intermittent exposure to high ozone concentrations may be the most damaging to plants (Sanz and Mil lan 2000; Krupa and Kickert 1997). Ambient ozone, present in concentrations of greater than 50ppb, is thought to adversely affect forest health (Reich and Amundson 1985). This may put the L F V forests at risk of oxidant damage due to sununertime ozone episodes (McKendry and Lundgren 2000; McKendry 1994). These episodes adversely effect local air quality and may cause ozone levels to exceed standards (Pryor et al. 1995). In Canada the national ambient air quality objective for hourly average ozone concentrations is 82ppb and has been set to protect human health (Stull 200). Currently, there are no air quality standards in Canada specifically for the protection of vegetation. European (AOT40) and American (SUM06) indices have been formulated to protect vegetation from ozone injury. They were developed by setting a maximum threshold at 40ppbv/h and 60ppbv/h respectively, and then by summing all cumulative exceedances of the of the threshold over a growing season or year. However, these indices were designed to protect crop plants and were tested / formulated under chamber conditions making them lose their value once atmospheric turbulence is considered (Krupa and Kickert 1997). Also, these indices are based on hourly averages and total exceedance values, when it has been shown that plants are particularly sensitive to short 'spikes' in ambient ozone levels (Sanz and Mil lan 2000; Krupa and Kickert 1997). Moderate concentrations of ozone over a long time period, may be 16 just as damaging to vegetation as intermittent episodes (McLaughlin 1998) and forest tree species may integrate exposure from growing season to growing season (Langebartels et al. 1998). In addition, these indices fail to take into account the recovery of plants during low exposure periods or the timing of exposure in terms of physiological susceptibility (Krupa and Kickert 1997). It is therefore uncertain whether these indices are sufficient to protect native vegetation growing in natural, turbulent, and variable environments. Despite the aforementioned limitations, these indices will be used in combination with the Canadian National Ambient A i r Quality Objective of 82ppb to examine data and deteniiine if there is a risk to forests posed by ozone in the LFV . A weighted metric (W126) that includes the role of recovery periods in plant injury but still excludes timing, has also been developed (Lefohn and Runeckles 1987; Griinhage andjager 1994), but will not be utilised in this study due to lack of detoxification rates for the vegetation of interest. 17 3.0 M E T H O D S Data collection involved collaboration with the Canadian Forest Service as a continuation of work started by 'Pacific 2001' under the 'Georgia Basin Ecosystem Initiative' project Pacific 2001 included collaborators from Environment Canada, the British Columbia Ministry of Parks, the University of British Columbia (UBC) and the Malcom Knapp Research Forest (MKRF). One of the project goals was to gain an understanding of how forests of the Lower Fraser Valley (LFV) are responding to ozone exposure (Stone 2001). 3.1 Study Area and Site Characteristics The L F V experiences a Pacific maritime climate with low-pressure systems bringing mild and wet weather during winter months, and high-pressure systems that bring warm and dry conditions during the summer. The area is characterised by complex coastal terrain leading to strong vertical and horizontal gradients in temperature, pressure and moisture. Ozone and meteorological data were collected at four sites in and near the M K R F (affiliated with the University of British Columbia) in the Municipality of Maple Ridge, less than 75 km east of downtown Vancouver, British Columbia. The location of the study area is shown in Figure 3.1. Data were collected from June 25'" to the 3 r f of October 2001 (Julian Days (JD) 176 - 276) and from June 21" to the 11* of October 2002 (JD 172 - 284). This study also includes ozone and meteorological data, collected by the M W A L P (Ministry of Water, Land and A i r Protection), from the municipalities of Pitt Meadows, Maple Ridge and Abbottsford, and from Burnaby Mountain, for the 2001 and 2002 study periods ( M W L A P 2003). The research-forest is on the north side of the LFV , situated in the foothills of the Coast Mountains. The northeast and eastern edges of the forest are bordered by the mountains of 18 Figure 3.1- Location of the study area in die L FV (inset), and die location of the four field sites and the tower (T).The descriptions of site 1 (200m), site 2 (400m),sitc 3(600m) and site 4 (120()m)are discussed within the text. Source: McKendry and Lungren 2001. 19 Golden Ears provincial park, to die west of the research forest is Pitt Lake, and to the south is the urban area of Maple Ridge (UBC Forestry 2001). In order for diurnal and altitudinal variation in ozone concentration to be examined, four ozone and weather monitoring stations were installed/deployed at approximately 200, 400, 600 and 1200 metres in elevation. Figure 3.1 shows the locations of each of the four study sites within the research area. Site 1 (200 m) was located at the entrance to the research forest at the U B C office in the 2001 season. Site 1 was moved to a more remote location for the 2002 study due to a potential interference in ambient ozone levels caused by automobile emissions from traffic to and from the forest Both locations were not in the forest itself, but in clearings beside a few small buildings. The location of site 1 in the 2002 study period was N49°16'26.2" W122°34'39.5" at an elevation of 188 m. The location of site 2 (396 m, N49°17'16.9" W122°33'22.4") was slighdy northeast of site 1 in a small forest clearing on a slope of the Spring River valley. Site 3 (588 m, N49°20'46.7" W122°33'58.2"), was located at the end of road H50 near the southern edge of Golden Ears Park, on a rocky ridge extending out from the forest Site 4 (1221 m, N49°20'00.6" W122°30'01.3") had a location on the southwest side of Alouette mountain in Golden Ears P.P., on a sub-alpine rocky ridge. The sites used were chosen by researchers from the Canadian forest service for use in the Pacific 2001 project, which included the 2001 ozone monitoring data used in this study. Sites were chosen for their accessibility except for the 1200m site, which requires helicopter access or a long hike. A l l site locations were required to be in small forest clearings, rather than in the forest canopy, due to a reliance on direct sunlight to feed the solar panels that ran the equipment (Humphreys 2001pers. com.; Unger 2001pers. com.). 20 Vegetation sampling for symptoms of ozone injury was conducted at each of the four continuous monitoring stations and the arboretum (near the 2001 site 1). This allowed visible injury and ozone exposures to be correlated quantitatively. Also, the sites used for monitoring plant injury were located in clearings, since the plant's leaf cuticle needs exposure to direct sunlight in order to display visible symptoms (Bergmann et al. 1999; Innes 2002 pers. comm). The arboretum had a mix of native and non-indigenous vegetation, mosdy planted. The dominant trees site 1 were Alnus rubra (Red alder), with scattered Acer macrofolium (Big leaf maple), Pseudotsuga menziesii ssp. menziessi (Douglas fir), and Thuja plicata (Western red cedar). Dominant shrubs consisted of Rubus spp. (all sorts), Spirea douglasii ssp. douglasii (Hardhack), Vaccinium parvifolium (Red huckleberry), and Pteridium aquilinum (Bracken fern) and the groundcover were gravel, grasses and Taraxacum officinale (Dandelions). Site 2 was characterised by Pseudotsuga menziesii ssp. menziessi (Douglas fir), Betula papyrifera (Paper birch), Alnus rubra (Red alder), Tsuga heterophylla (Western hemlock), and Thuja plicata (Western red cedar), and the shrubs Gaultheria shallon (Salal), and Vaccinium parvifolium (Red huckleberry). The ground at site 2 was mosdy bare rock with some sparse moss colonisation. The forest surrounding site 3 was composed mainly of Thuja plicata (Western red cedar), and Tsuga heterophylla (Western hemlock), with Gaultheria shallon (Salal), Vaccinium parvifolium (Red huckleberry), and Pteridium aquilinum (Bracken fern) composing the under-story shrub layer. Site 4 was sub-alpine with Tsuga mertensiana (Mountain hemlock), Abies amabilis (Amabilis fir), and Chamaecyparis nootkatensis (Yellow-cedar) as scattered krumrnholz and Vaccinium ovalifolium (Oval-leaved blueberry), V. membraceum (Black huckleberry), Cassiope mertensiana (White mountain heather), and 21 Phyllodoce empetriformis (Pink mountain heather) as dominant shrubs. The ground at site 4 was mosdy bare or lichen covered rock. The tower was located just south of site 3 (546masl N49°20T2.8" W122°34'00.2") off the road in a narrow clearing and is indicated by the "T" in Figure 3.1. A clearing was required so that the tower could be assembled on the ground prior to erection. A n area of low trees was chosen due to limited tower height (10.5m) so that measurements would best reflect gradients of ozone concentrations seen within an entire forest canopy, from top to bottom. Tree height in the area ranged between 3m and 13m. Dominant vegetation included the larger Thuja plicata (Western red cedar) and Tsuga heterophylla (Western Hemlock). A short but dense stand of young Alnus rubra (red alder) was located beside the tower filling much of the clearing. Shrubs and tree seedlings filled the under-story of the adjacent closed canopy forest. 3.2 Continuous Monitoring for Altitudinal Variation Continuous monitoring of ground level ozone began in June of 2001 and 2002, with solar powered ozone monitors and adjacent weather stations recording hourly average; ozone levels (ppb - parts per bill ion by volume); high and low temperatures (°C); wind-chill (°C); mean and high wind-speed (m/s); and wind direction (bearing). The measuring of ozone and weather variables, ended for both the 2001 and 2002 season in early October since, autumn and winter in the L F V bring rain and cloud of sufficient intensity and duration that solar panels are virtually useless. Also, site 4 becomes inaccessible due to snow. Weather and ozone data from nearby sites in the region were obtained from the M W L A P . Each site was equipped with a 250cm tripod for instrument mounting, a meteorological package, ozone-monitor and power supply. The met-package included; a Davis EZ-Mount 22 Weather Wizard III (in a locked box at approx. 150cm above ground) that initiated measurements and stored the data supplied by a wind vane; cup-anenometer; and temperature/relative humidity sensor. The wind instruments were mounted at the tripod's top, and the sensors (with covers) were mounted between 50cm and 200cm in height The 'weather wizards' made measurements every 30 seconds and then averaged the measurements for each hour. The ozone monitors were from '2B Technologies' in Golden Colorado and measured ozone using ultra-violet light absorption at 254 nm. Measurements of ozone concentration were taken every 10 seconds and then averaged for each hour. They recorded hourly ozone averages in parts per-billion (ppb) for the two monitoring periods through a closed-path tube and filter system, at 100cm from the ground. The monitors were in locked pelican cases beneath the tripod. Three solar panels facing south were mounted to the top, middle and bottom of the tripod, tilted upwards. The panels supplied three 10V car batteries in locked pelican cases beneath the 2B-montior. The readings were stored by logging devices built-in to the equipment, and the apparatus set-up is shown in figure 3.2. In 2001, only sites 3 and 4 were equipped with a 'weather wizard'. The weather data for site 1 came from Environment Canada researchers working in the M K R F office, and the site 2 data was obtained from Michael Feller in the department of Forestry at U B C whose fire weather station shared a location with site 2. Unfortunately the fire station only measured weather variable once a day - at 12 OOh. There are also some gaps in the 2001 ozone and met data, as only one solar panel was used for each station at the beginning of that monitoring season. Cloud coverage for extended periods of time meant that power was not supplied to the stations and no data were collected. Gaps in data increase with site elevation due to increases in cloud cover. A second solar panel was installed at each of the four stations in Sept2001 and a 23 Met-one wind vane and cup anenometer Temperature and relative humjcity sensor Figure 3.2 - The site set-up used for continuous monitoring of hourly ozone concentration and meteorological variables, for both the 2001 and 2002 study periods. third in 2002. There were no losses of power experienced by any of the monitoring stations during the 2002 season. In the 2001 study period, the main hmitation of the ozone monitors was the lack of a 'real time' clock. Rather than hourly averages, data was logged based on timers set to every 57+ minutes and these readings were adjusted to fit the weather data to the nearest hour in 2001. Real time clocks were installed by 2B-technologies in preparation of the 2002-study period. The 2002 season had large data gaps independent of power supply. One-month of continuous ozone data were lost at sites 2 and 3 spanning Aug.l9-Sept 17 due to logging < error. 24 Sites 1 and 4, at the end of the season, received no ozone measurements when U V bulbs burnt out (from August 27 at Site 4, and August 19 at Site 1). Smaller gaps in the 2002 continuous data are caused by the need to turn off the 2B monitors before logging if there was a blank screen. Each time this was necessary, up to ten hours of data were lost. Also, it seems that the 2B monitor at site 2 (400m) was malfunctioning for the entire 2002 study season due to filter contamination. For this reason all of the site 2 data for 2002 was discarded and will not be analysed in this study. There were no gaps in the 2002 meteorological data set and all four sites were equipped with 'weather wizards', which recorded variables as hourly averages. Data were downloaded from the respective logging devices once every two weeks using a laptop computer and hyper-terminal, on every other Tuesday, alternating between the three lower sites (1,2 and 3) and site 4 which took many hours to reach. Downloading every two weeks also minimised data loss due to powering-off the monitors. 3.3 Passive Monitoring for With in Canopy Variation During the 2002 field season a 10.5 m tower was designed, and erected in a small and narrow forest clearing in order for tropospheric ozone profiles to be examined at a smaller scale. The tower design is shown in figures 3.3a and 3.3b. The tower was constructed of three aluminium poles of 4 m, 3 m and 3 m in length from bottom to top, respectively. Three Vaisala relative humidity and temperature sensors were attached to the tower with weather covers at heights of 0.36m, 4.65m and 9.60m - near the bottom, middle and top of the canopy respectively. Three aluminium bars of lesser diameter were attached perpendicular to the tower pole. One crossbar attached to an extension of the tower-top, supported a Met-one wind-vane and cup-anenometer. Measurements made by the Met-one and Vaisala instruments were processed and recorded using a Fischer Scientific 2 I X data-logger and a Pc208w software 25 package. Temperature, relative humidity, wind speed, wind direction and wind vector were recorded every 10 minutes and converted to one-hour averages for downloading. Meteorological data for the tower were lost from August 8 - 2 7 due to do downloading error using the Pc208w software. The other two cross-arms were attached to the top and bottom of the main tower with pulley wheels (4 in total) welded to their ends. Braided white synthetic rope of 1.5 cm diameter was run through the pulley system on either side of the tower, thus allowing for passive samplers to be attached at varying heights while being raised and lowered with ease (see figures 3.3a and 3.3b). Ogawa & Co. Inc. passive ozone samplers were used to monitor within canopy, as they^are light, inexpensive and have been proven to yield good results (Krupa et a/. 2003; Sather et al. 2001; Ray and Flores 2000; Brace 1996; Brauer and Brock 1995). Figure 3.4a) shows the Ogawa & Co. passive samplers, with components in order of assembly. The samplers are composed of a cylindrical sampler body and components - made of teflon - and at each end of the sampler body are paper filters coated with a nitrite-based solution, with a steel screen on either side. The sampler body is contained within a P V C clip to be attached to rain covers etc. Ambient ozone reacts with the nitrite in the filter to form nitrate and the ratio of these nitrogen oxides is then analysed to calculate cumulative ozone exposure in ppb over the sampling period. Cumulative ozone exposure is useful when examining vegetative injury since plants are subject to a cumulative dose over the growing season. 26 Figure 3.3a - The 10.5m tower erected in a forest canopy to measure meteorological varibles and cumulative ozone exposure with height. A l l components are shown with labels in the illustration in Figure 3.3b. 27 Figure 3.3b - The lower designed and creeled within a low forest to measure ozone extx>sure with height in the canopy. 28 Figure 3.4 - a) Ogawa & Co. Inc. ozone sampling badge composed of: 1. P V C clip; 2. teflon cylindrical sampler body; 3. teflon spacer disk; 4. teflon spacer ring; 5. steel screens; 6. nitrite coated filter; 7. teflon diffusion end-caps, b) Ogawa & Co. Inc. rain-covers made of P V C (approx. 9cm diameter) used at the continuous monitoring sites. Similar rain-covers were fashioned for use on the tower. 29 Rain covers were made for the passive samplers out of white P V C piping and pipe caps of 15cm diameter. The rain-covers were equipped with a cross wire inside to which the Ogawa passive samplers were be clipped. A similar rain cover design, made by Ogawa & Co, is shown in Figure 3.4 b). The wires were covered with fine teflon tubing to inhibit ozone deposition on their surface. The samplers hung at the standard 1.3 -1.5 cm from the bottom of the rain cover (Rupprecht& Patashnick Co. Inc. 2002; Ray andFlores 2000). Seven of these samplers with rain-covers were attached to each of the ropes on pulleys, using white Velcro straps and a sophisticated wrapping technique. The samplers were evenly spaced at a distance of 155 cm from one another. The two sides of the tower, or two pulley systems, acted as a replicate so that each sample period included fourteen passive samplers. Passive samplers recorded cumulative ozone exposure with height in the forest-canopy at two-week intervals. Five of these sampling periods occurred over the study, however one was of a twelve-day duration due to supply difficulty. Although the two-week averaging interval cannot lead to a temporally detailed analysis of within-canopy ozone variation, it is a satisfactory to obtain relative exposure values with height (Karlsson et al. 2002; Bytnerowicz et al. 2002). Passive sampler filters were analysed using ion chromatography following the Harvard School of Public Health protocol (Harvard School of Public Health 2001). The only change in the protocol was not using a glove box; instead the laboratory environment was kept clean and dust-free, and work was done quickly to limit the exposure time of the filters to ambient air in non-field conditions. A copy of the protocol used to analyse the filters is given in Appendix A . Three field blanks were used in each 2-week sampling periods (save the first sample set for which there were not enough filters). Also, each of the continuous monitoring sites was 30 equipped with a passive sampler and replicate, in Ogawa & Co. P V C rain covers, to correlate between continuous sampling with passive ozone sampling. Again, due to the low number of filters obtained for the first sample period there were no replicates - only one sampler - at each of the four continuous monitoring sites for the first passive monitoring period. Passive ozone monitoring began on the 9* of July and ended on the 1" of October, 2002. However, the period from July 23"" to August 8* has no passive data due to difficulties in ordering filters from C D Nova in Burnaby BC , and instead filters were ordered directly from Japan via Ogawa & Co. USA . The sampling period that began on August 8* lasted only 12 days (until the 20*) so that the samplers could be put out in the field as soon as possible, without rearranging the field-schedule. Samplers were changed every other Tuesday, alternating between the three lower sites (1,2 and 3) and tower; and site 4, which took many hours to reach. Met data were downloaded from the 2 I X using a laptop computer at the same time that the samplers were changed. 3.4 Characterisation of Ozone Indices Both AOT40 , and SUM06, were discussed in terms of utility and hmitation in the last chapter. The European (AOT40) and American (SUM06) indices were formulated, by Fuhrer and Acherman (1994) and Hogsett et al. (1995) respectively, to protect vegetation from ozone injury. They were designed primarily for crop protection but may be, and have been, utilised for native plant species' injury analysis. Due to the hmited number of chamberless field studies to develop indices for natural vegetation, these indices are generally applied in ambient field studies (Fuhrer er al. 1997). They are calculated by setting a maximum threshold at 40ppbv/h for A O T 4 0 and 60ppbv/h for SUM06, for hourly mean concentrations. Any exceedances of the threshold levels by hourly ozone concentrations are summed over the 31 growing season or year, in this case over the summer months. The A O T 4 0 is a cumulative sum of differences between 40ppb and actual concentrations >40ppb (e.g. 1 + 2 + ...n), whereas SUM06 represents the sum of all actual concentrations >60ppb (e.g. 61 + 62 + ...n). Both of these indices measure cumulative ozone exposure over die desired threshold for the season. The indices are presented as a cumulative time series to illustrate the cumulative nature of plant tissue exposure to ozone. A O T 4 0 and SUM06 were both calculated using data from all four study sites, for both 2001 and 2002 seasons. This study includes the exceedances of critical levels over a 24-hour period for each day, and includes all hours for which ozone data were available. Other studies have calculated cumulative indices based on 24-hour data (Ghosh eta/. 1998), although other use daylight hours only (VanderHeyden et ai. 2001; Karlsson et ai. 2002). Musselman and Minnick (2000) suggest the use of 24 hour data when calculating cumulative exposure indices as vegetation may be particularly susceptible to ozone injury at night 3.5 Vegetation Sampling for Foliar Injury A t the end of July all vegetation at the four continuous sampling sites was examined for any discoloration to foliage that may be indicative of ozone injury. The end of July was chosen since it is late enough in the growing season that leaves have fully matured, but early enough in the season to eliminate autumnal senescence as a cause of discoloration. Symptoms considered indicative of potential injury caused by ozone exposure included; upper surface leaf reddening, or purpling - often as a stipple - that was absent on the underside of the leaf and the leaf veins, both of which should remain green. Plants were also included in the study if they showed inter-veinal necrotic stipple or chlorosis (Hogsett et ai. 1985; Kickert and Krupa 1997; Bergmann et 32 al. 1999; Brace et al. 1999; Innes et al. 2001). A photo of foliage from Cornus sanguinea displaying 'typical' ozone injury is given in figure 3.5. Due to a lack of experimental information on ozone injury under natural conditions, this study uses the results of chamber fumigated and filtered air studies to assess ozone injury under ambient field conditions, as suggested by Kickert and Krupa (1991). Despite limitations in verifying injury Bergmann et al. (1999) suggest visible injury as not only the first sign of damaging exposures, but also the most reliable way to monitor high ozone levels. Figure 3.5 - Typical injury ozone injury symptoms on Comus sanguinea following controlled chamber experiments. Source: Innes el ai. 2001 Individual broad-leaved shrubs (with the exception of a Douglas-fir sapling at site 2) displaying any of the symptoms mentioned above, were chosen at each of the four sites and at the arboretum. Shrubs, being low-level vegetation, made it possible to examine all foliage on an individual. Fast-growing shrubs may also be more sensitive to ozone damage due to 33 increased uptake caused by a high stomatal conductance (Reich and Amundson 1985), and may prove useful as bioindicators. Table 3.1 gives a summary of the individuals chosen, their species, and site location. For the site 4 species in Table 3.1, both V. membraceum and V. parvifolium represent three monitored individuals. Once these plants were chosen as subjects they were tagged and returned to on a biweekly basis at which time they were photo-graphed using a digital camera to document injury symptoms and their development, or lack of development, over time. Brace et al. (1999) urge researchers to take photographs of individuals when attempting to assess injury under ambient field conditions. A l l assessments were made by the same observer to ehminate invalid results based on personal bias or opinion. The Horsfall-Barratt (HB) scale was used to determine the degree of injury (HorsfaU-Baratt 1945) and to determine whether or not symptoms worsen over time. The Horsfall-Barratt scale combines estimates of the percent of the entire plant having injured leaves in 5% increments (0, 5, 10, 15 ... 100%), and the percent of leaf area injury. The leaf area scale (0, 1 , 3, 6, 12, 25, 50, 75, 88, 94, 97, 99, 100%) is more precise at the high and low ends to help eliminate error associated with mid-range damage estimates (Ghosh et al. 1998; Innes et al. 2001). The percentages are then multiplied to give an overall injury rating that may be compared between species', location, and for an individual over time. Since the H B scale is based on the product of two percentages, values obtained from the injury score may be as high as 10 000 for an individual expressing 100% leaf area injury on 100% of leaves. . 34 § Ai bou turn Spirea clouglasi Hardback « Vaazniiini IMIWlolllllIl ' ^Red^Huckleberry Site>l - > i*f t0"# * '^ Rubus pamfohus Thimbleberry \ Itanium j parviiokum Red IlmkJcli«.ir\ Siii 2 •PsuedotsiigiL: menziesii Douglas-fir iSite ^ Gaultheria shallon Salal ' M i l 4 memtiaceunj Black Huckleberry* Gaultheria shallon V parvifolium ' Vaccinium ovalifolium SalaJ wgmsmm Red Huckleberry V parvifolium V membraceum Red Hucklebeiry Black I Iucklebcny Acer circmatum ' Acerglobrum ' Vine Maple Rocky Mountain Corn ws canadensis Dwaif Dogwood Otnl-lcucil Bluebi 11 \ ' WM6 Table 3.1 - Species, and their common names, that were chosen for the monitoring of visible ozone injury over the 2002 study period listed by site location. * Indicates three individuals of that species were monitored. 35 4.0 A L T I T U D I N A L A N D SEASONAL V A R I A T I O N I N O Z O N E Ozone and meteorological data were collected at four sites of varying elevation located at approximately 200, 400, 600 and 1200 m.a.s.l. The complete time series' of ozone concentration and meteorological variables are given for both the summers of 2001 and 2002 in Appendix B. The data are separated into two-week periods and periods of data absence are also included so that the figures are equally scaled. The following is a discussion of how tropospheric ozone varies with elevation, time of day, synoptic condition and distance from the urban source in the Lower Fraser Valley (LFV), British Columbia. 4.1 Altitudinal Comparison Figure 4.1 is a box and whisker plot of median, maximum, minimum, and the first and third quartiles (Inter Quartile Range) of ozone concentration at each of the four sites for all hours of the 2001 study period. The plots in the figure show the median, rather than the mean ozone concentration. This is valuable because ozone concentrations do not reflect a normal distribution in a time series, but instead hourly averages show a marked skew to the right (to higher concentrations) (Kickert and Krupa 1991). Maximum concentrations are important to note since intermittent exposure to high ozone concentrations may be the most damaging to plants (Sanz and Mi l lan 2000; Krupa and Kickert 1997). The data used to calculate these variables for 2001 include all data collected for the 100 day period from June 25* - October 3"1, 2001 (n=2395) for sites 1, 2, 3, and 4 at approximately 200m, 400m, 600m, and 1200m elevation, respectively. In general ozone increased with elevation above 400m. Site 1 at 200m had seasonal median ozone concentrations exceeding those at 600m. However, this site was located at the entrance of the research forest beside a 36 parking lot where local automobile emissions may increase ambient ozone. For this reason, site 1 does not represent the site of lowest mean concentration as would be expected by its elevation. Recall that the location of site 1 (200m) was changed from 2001 to 2002. Also recall that site 2 at 400m was removed from this data set due to instrument error throughout the season. Figure 4.2 is a box and whisker plot of median, maximum, minimum, and the first and third quartiles of ozone concentration for each of the four sites in 2002. The data used to calculate these variables include all data collected for the 109 day period from June 21" -October 8*, 2002 for sites 1, 3, and 4 at approximately 200m, 600m, and 1200m elevation, respectively. Median summertime ozone concentration increased with elevation in 2002 as it did in 2001. Site 1 at 200m had maximum ozone concentrations exceeding those at both 600m and 1200m, but had a lower median concentration due to the stronger diurnal variation seen at lower elevations. Figure 4.1 - Box and whisker plot summarising 200l's altitudinal variation in ozone concentration. 37 A complete summary of general ozone variation by site is given in Table 4.1 for 2001 and in Table 4.2 for 2002. The information includes mean concentrations, deviations and spread for all hours of the study season. Site 1 had a mean ozone concentration of 18.13ppb for the summer of 2001 - less than 3ppb higher than ozone averages for sites 2 and 3. However, site 4 had a mean \ozone concentration of 32.40ppb more than twice as high as the mean ozone concentration at sites 2 and 3 representing ozone concentrations at lower elevations. The highest elevation site, site 4, also had the highest maximum concentration of 87.86ppb, almost 6pbb above the National Ambient A i r Quality Objective (NAAQO) of 82ppb for ozone in Canada. It is clear that site 4 at 1200m experiences much higher ozone concentrations on average than the lower elevation sites. It was found that the seasonal maximum values for all four of the sites, occurred during the period from August l l l h - August 15*, 2001, an episode to be discussed later in this chapter. Station 200m 400m 600m 1200m Mean 18.13 15.20 15.95 32.40 Standard Deviation 12.67 12.45 , 12.03 14.10 Maximum 80.32 72.34 71.33 87.86 Min imum 0.00 0.00 0.00 0.00 Inter Quartile Range 14.84 15.02 15.29 16.66 Table 4.1: Summary of ozone concentrations for eac i site in the summer of 2001 Figure 4.2 shows that in contrast with 2001, the lowest site at 200 m.a.s.l. had the highest seasonal maximum concentration of 96ppb in 2002 - 14pbb above the N A A Q O of 82ppb for ozone in Canada. Site 1 exceeded the 82ppb threshold for three hours during the 2002 study-period. Although sunmiertime average ozone exposure seems to increase with site 38 elevation, all sites had a low minimum value of Oppb or lppb. This shows again that even at high elevation sites ozone is on occasion effectively depleted. Site 1 at 200m experiences much higher peaks in ozone concentration during the 2002-study season than the other two sites, but has a lower mean ozone exposure due to low night-time concentrations. Site 4 at 1200m experiences the highest ozone concentrations on average but has a lower maximum concentration implying weaker peak ozone levels. Figure 4.1 - Box and whisker plot summarising 2002's altitudinal variation in ozone concentration. The standard deviation and thus variation in mean ozone concentration was high at all sites and is most closely related to peak concentrations since in both years all sites experience depletion. The differences in mean concentrations between sites are due to the strong diurnal cycle of ozone concentration, which will be discussed in the following section . 39 Station 200m 600m 1200m Mean 12.97 19.02 21.88 Standard Deviation 14.23 10.64 12.09 Maximum 96.00 65.00 69.00 Min imum 0.00 1.00 1.00 Inter Quartile Range 16.00 13.00 17.00 Table 4.2: Summary of ozone concentrations for each site in the summer of 2002 4.2 DiurhaJ trends Figure 4.3 shows seasonal means, by hour of the day, of ozone concentration at all four 2001 continuous monitoring sites for the entire study period (n=100). Similar graphs of mean diurnal variations in ozone concentration for the growing season, at differing elevations are given in Krupa and Kickert (1997). Ozone levels are the lowest at all sites throughout the night-time hours (after 23:00 PST) and begin to increase around 8:00 PST as the sun rises promoting both the photolysis of NO* and the convective down-mixing of elevated layers. Ozone concentration reaches its peak between 16:00PST and 19:00 PST at all four sites. During the day-time (8:00-19:00 PST) ozone levels increase with elevation, with the exception of site 4 (1200m) where ozone levels remain high throughout the diurnal cycle. During the night-time hours (21:00 - 4:00 PST) ozone concentrations clearly decrease with elevation. The 'in between' times (4:00-8:00 PST and 19:00-21:00 PST) can be considered times of cross-over, or trend reversal, in the ozone profiles. Figure 4.4 shows seasonal means of ozone concentration at all three sites included in the 2002 study period (n=109). Diurnal patterns are similar to those in 2001 mentioned above, with ozone levels being the lowest at all sites throughout the night-time hours (after 23:00 PST) and beginning to increase around 8:00 PST. Mean ozone concentration reaches its peak between 16:00 PST and 19:00 PST at all sites as it did in 2001, due to the travel time required 40 15 * §8888888888881888 8 8 8 8 8 8 8 i - l i - H i - H r - t r - l p H i - l r - l r H i - l p t O * ? * ? * Time Figure 4.3 - Mean hourly averages of ozone concentration at sites 1 ,2 ,3 and 4 for the entire 2001 monitoring period. Error bars represent onestandard deviation from the mean. for the urban plume to reach rural areas, creating a time lag between peak production, and peak ozone concentration (Brace and Peterson 1998). In areas nearer to urban sources ozone reaches an early afternoon peak between 12:00 - 15:00 PST (Kickert and Krupa 1997). A morning increase (McKendry and Lundgren 2000) and night-time decline in ozone levels (Giisten etal. 1998; Banta etal. 1997; Pisano etal. 1997; Fast and Zhong 1995) are consistent with previous findings of diurnal ozone variation. Diurnal variations in ozone concentration decrease with altitude in both years leading to the increase in mean concentrations with altitude discussed previously. Site 1 at 200m has the greatest diurnal variation and site 4 the least in 2001, while in 2002 site 3 and 4 exhibit similar diurnal trends to one another. The lack of diurnal variation at sites 3 and 4 in 2002, leads to theses sites experiencing higher mean concentrations as mentioned in section 4.1. Site 4, the highest elevation site, has higher average ozone concentrations throughout the night (on 4 1 average >16ppb) than the lower sites which have night-time average concentrations of < l lppb. This difference in seasonal means between sites is large enough to be considered significant. Ozone is depleted to a much lesser extent at high elevation sites, and this leads to the nocturnal persistence of ozone at higher elevations. Nocturnal ozone depletion appears to be a much more important process controlling concentration at low elevation sites, than at higher elevations (Zaveri et al. 1995; Brace and Peterson 1998; and Cooper and Peterson2000). Time « « « Figure 4.4 - Mean hourly averages of ozone concentration at sites 1, 3 and 4 for the entire 2002 monitoring period. Error bars represent one standard deviation from the mean. Peak concentrations decrease in magnitude with height up to 800m, but site 4 (1200m) has the highest peak levels on average and the highest ozone concentration throughout the day for the 2001 study period, whereas site 1 had the highest peak concentrations in 2001. It is these relatively high ozone exposures and a lack of diurnal variation, that leads to these sites experiencing the higher mean concentrations as mentioned in section 4.1. Peak concentrations are on average the highest at site 1 - the lowest elevation. Even though site 1 experiences a 42 relatively high peak ozone exposures, the strong diurnal variation (nocturnal depletion) keeps its seasonal mean low. Sites 3 and 4 exhibited much more similar diurnal trends in 2002 than in 2001. This is suspected to be due to seasonal differences in boundary layer development, but due to an absence of temperature soundings in the region that hypothesis can not be tested. It is thought that more subsidence and a shallow boundary layer in 2002 may have brought/concentrated more pollutants at lower elevations (600m) while often failing to intercept site 4. The error bars in Figures 4.3 and 4.4 represent one standard deviation from the mean. The extent of error bar overlap shows that differences between sites are statistically significant. Seasonal hourly averages differ between sites of various elevations. Patterns of deviation from the mean may be considered representative of the sites for the 2001 season, with seasonal standard deviations increasing with altitude and increasing towards solar noon when all sites show similar deviations. Standard deviations would be the highest for daylight hours when temperature, solar radiation and other factors leading to ozone formation are the most variable. The night-time concentrations of ozone at high elevations are largely controlled by the levels that existed in the previous day's mixed layer, while lower elevations always experience nocturnal depletion keeping the low elevation standard deviations low at night 4.3 Synoptic Influence on Episodic Ozone Increases Ozone and meteorological data were obtained from the B C Ministry of Water, Land and A i r Protection (MWLAP ) for Abbottsford, Burnaby Mountain, Maple Ridge, and Pitt Meadows from the web-site (http://www.elp.gov.bc.ca:8000/pls/aqiis/aqi.bulletin). These selected areas / municipalities are all located in the L F V between the city of Vancouver and the 43 study area, and are sites where ozone is continuously measured as hourly averages. A l l of the M W L A P monitoring stations are located at elevations at, or near, sea level. Surface pressure and 500mb geopotential height maps of the region were obtained from the web-site of N O AA -CIRES Climate Diagnostics Center in Boulder, Colorado U S A (http://www.cdc.noaa.gov/). The following is a discussion and analysis of episodic ozone increases in the LFV . 'Episodes', for the purpose of this study, are defined as periods when ozone concentrations are above average values for much of the day. Two such events will be examined in detail and will include days before and after the actual event The periods to be examined are August 8* -17V2001 (200-229) andjuly 19* - 28*, 2002. The discussion of both events will include hourly ozone and temperature data from the field sites as well as from the M W L A P sites. Each sub-section will conclude with a discussion of synoptic circulation (sea level pressure, and 500mb geopotential heighO over region during the episodes. 4.3.1 2001 Episode August 8-17 The period from August 8* to August 17* 2001 was characterised by tropospheric ozone concentrations much higher than seasonal and hourly means. These nine days also mark the time when each of the four study sites reached their maximum concentrations for the 2001 study period (See Table 4.1). A time series of average hourly ozone concentration at each of the four sites and hourly average temperature at site 1 (the nearest to sea level) is shown in figure 4.5. For the following discussion, August 11* - 15* (JD 223-227) will be discussed in terms of the 'episode' with the other days defining the period leading to the build-up and the decrease, of ozone in the region. 44 Julian Day Figure 4.5 - Hourly average ozone concentrations at the four study sites for the August 8-17, 2001 episode. Hourly average temperature at site 1 (200m) is also included. Hours when mean ozone concentrations met or exceeded the N A A Q O are labelled. This episode occurred in conjunction with very warm midday temperatures in exceedance of 25°C from August 8* - 15*. Temperatures decreased towards the end of the episode. Temperature is strongly correlated with increases in ozone production due to the temperature dependency of the reactions that produce ozone (Sillman 1993) and the necessity of high intensity solar radiation to facilitate the photolysis to nitrate (Oke 1987). However, the daily maximum in ozone concentration occurs later in the day than the temperature maximum. This is because the formation of ozone occurs closer to the city and the urban plume must travel to reach the study area eastward in the valley. This time lag in peak ozone concentration with increased elevation since the pollutants are carried up-slope by daytime anabatic winds. Also, there appears to be a lag of nearly a day between the end of the episode for site 1 through 3, and the end of the episode for site 4. Notice in figure 4.5 that ozone at site 4 remains very 45 high on August 16* (JD 228) and into the early morning of August 18*, a time when concentrations were back to near average values at the other three sites. Peak ozone concentrations are the highest at site 4, the site of highest elevation. A t site 4, during the episode in August of 2001 levels exceeded N A A Q O of 82ppb, for five hours in four, out of five 'episode' days (Aug. 11-14). None of the other three sites exceeded the 82ppb threshold throughout the study period. The detailed time series of the episode shown in figure 4.5 further illustrates the decrease in diurnal variation with increased elevation that was discussed in terms of seasonal hourly averages in the previous section. A t site 1 ozone concentrations are high throughout the late afternoon (>50ppb) for the five days, but drop down to <10ppb on all days but one (Aug. 11 J D 224), including a complete ozone depletion to Oppb in the early morning hours towards the end of the episode. Sites 2 and 3, at 400m and 600m elevation respectively, exhibit peak concentrations for the five days, in exceedance of 50ppb. However, in the early morning minimum ozone values at site 2 are generally on the order of lOppb lower than at the higher elevation, site 3 with minimum levels around 30 -20pbb. Site 4 exhibits little diurnal variation due to a lack of night-time depletion. Min imum values at site 4 are greater than 40ppb for August 11-14 (JD 223-226) and occur in the late afternoon, a time when the other sites are experiencing their peak concentrations. The highest ozone concentrations at site 4 occur near midnight with a secondary peak in the early morning, when ozone at the lower elevation sites is at a minimum. Aneja et al. (1994) suggest that as elevation increases the diurnal cycle becomes less discernible, and may even reverse to show a night-time peak as seen here in the LFV . 46 The ozone episode described above also affected other parts of the L F V including municipalities between the study site and Vancouver. Figure 4.6 gives a time series of hourly ozone averages for August 8 - 17, 2001 at Burnaby Mountain, Pitt Meadows, Maple Ridge and Abbotsford, along with hourly average temperature at Abbotsford. Ozone concentrations in excess of 60ppb occurred at each of the sites throughout the episode and at one of the locations, Maple Ridge - the nearest of all to the four forest sites, uhe N A A Q O of 82ppb was exceeded for one hour {13:00 PST) on the 14* (JD 226). - Bumaby Mountain •Pitt Meadows Maple Ridge - Abbotsford Julian Day Figure 4.6 - Regional hourly average ozone concentrations at Burnaby Mountain, Pitt Meadows, Maple Ridge and Abbottsford for the August 8-17, 2001 episode. Hourly average temperature at Abbottsford is also included. Hours when mean ozone concentrations met or exceeded the N A A Q O are labelled. The suburban sites exhibit a diurnal ozone pattern much more closely related to temperature than the forest sites. This is due to direct emissions of ozone precursors near the monitoring stations causing ozone to be produced at the times of most intense solar heating, 47 and to be measured upon production. Sources of ozone are limited in the forest environment compared to the urban environment, even though forests emit some ozone precursors. The Vancouver plume is apparent in this data, by the time lag of maximum ozone concentration at sites further from the city. Abbottsford and Bumaby mountain, which are nearer to Vancouver, show high ozone levels beginning on the 10* (JD 222), whereas concentrations Pitt Meadows and Maple Ridge do not increase as quickly. The two more rural locations further from the city, exceed the other two sites in peak concentrations by the 12* (224). In addition, at the Burnaby Mountain location, of a higher elevation than the other three stations, exhibited a secondary night-time peak in ozone concentration similar to that seen at site 4 for the same period. This secondary nocturnal maximum occurs when there is a temporary turbulent coupling of the residual layer to the surface (Salmond and McKendry 2002). Nocturnal cooling leads to strong katabatic winds that may also bring pollutants from the residual layer down mountain slopes, unless the site intercepts the R L directly (Zaveri et al. 1995; Brace and Peterson 1998).. The August 2001 episode was dominated by synoptic events characteristic of above average ozone concentrations. Mean sea level (MSL) pressure maps showing the region from 40° - 60°N and 135° - 100°W are given in figure 4.7. Maps are given for the 8* (JD 220) figure 4.7a), 12* (JD 224) figure 4.7b) and 17* (JD 229) figure 4.7c) of August, 2001. The maps show the sub-tropical high pushing into British Columbia from the southwest and the development of a thermal trough along the west coast extending northward from the United States. Both of these surface features are associated with above average ozone concentrations (McKendry 1994). The thermal trough is thought to inhibit the sea breeze with the advection of warm air by south-westerly flow on the upstream side of the ridge. Sea breeze weakening leads to 48 a) b) c) August 8, 2001 August 12, 2001 m U 2 * ! M * 1 2 6 * 1 2 J * 1 2 0 * 1 ( 7 * 1 1 4 * ? t l * 1 0 8 * 1 0 5 * 1 0 2 * August 17, 2001 135* U J * 1 2 « * 1 2 6 * 1 2 2 * I S O * ' 1 7 * 1 1 4 * 1 1 1 * 1 0 8 * 1 0 S * 1 0 2 * Figure 4.7 - One day means of sea level pressure in millibars (mb), lor August 8th, 12th and 17th, 2001. Source: NOAA-CIRES Climate Diagnostics Center, Boulder, Colorado, USA. From their website http://www.cdc.noa.gov/ 49 a) 1)) August 8,2001 August 12, 2001 (39» UJW 1]W 12J» 130W I17W 11 • » I t l W 1 C 8 W 1 0 5 * I 0 J W C) August 17,2001 1 J » M2* 1?9«t 1Z6W 12W 1J0H 117W I14W U K 10d* 105» 102» Figure 4.8 - One clay means of500mb geojxrtential height in decametres, for August 8th, 12th and 17th, 2001. Source NOAA-CIRES Climate Diagnostics Center, Boulder, Colorado, USA. From their website http://vnvw.cdc.noa.gov/ 50 stagnation allowing for ozone to build up in the western portion of the L F V during these conditions. These same conditions are responsible for the high daytime temperatures that were recorded, clear skies and reduced mixing depths leading to enhanced ozone formation and / or persistence (McKendry 1994). A t 500mb for the same area and time period (figure 4.8) an upper level ridge developed east of 120°W; a synoptic feature most commonly associated with ozone episodes above and beyond 60ppb (McKendry 1994). The strength of the ridge is the greatest during peak ozone levels and as the ridge weakens ozone levels decline. In general the highest concentrations of ozone occur when this upper level ridge is combined with the high surface pressure and thermal trough along the coast (McKendry 1994). The combination of these surface and synoptic patterns is responsible for the ozone episode in the L F V from the 11* -14* of August Another period of increased hourly average ozone concentrations during the summer of 2001 occurred earlier in the summer between July 1 - 6, 2001 (JD 182-187). This is illustrated by the time series data presented in Appendix B . l . It is similar in extent and severity to the episodes recorded in 2002. 4.3.2 2002 Episode July 19-28 The period from July 19 * to July 28 * 2002 was characterised by tropospheric ozone concentrations higher than seasonal and hourly means (See Table 4.2), however levels do not reach those seen during the August 2001 episode described above. In addition, churnal and altitudinal trends are not as well established as in the above-mentioned event The time series of average hourly ozone concentration at each of the three sites and hourly average 51 temperature at site 1 (the nearest to sea level) is shown in figure 4.9. For the following discussion, July 19- 26 (JD 200-209) will be discussed in terms of the 'episode' with the later days defining the period of ozone decreasing to near average values. This episode occurred in conjunction with very warm midday temperatures often in exceedance of 30°C from July 20 - 23. These midday highs decreased towards the end of the episode. However, the daily maximum in ozone concentration occurs a few hours later in the day than the temperature maximum due to the travel time required from the plume source, discussed previously for the August 2001 episode Julian Day Figure 4.9 - Hourly average ozone concentrations at the four study sites for the July 15-28, 2002 episode. Hourly average temperature at site 1 (200m) is also included. Peak ozone concentrations are not the highest at site 4 (1200m), as they were during the 2001 episode. During the episode in August of 2001, levels at site 4 exceeded the N A A Q O of 82ppb, for five hours. Conversely, the July 2002 episode marked a time when none of the 52 sites had ozone concentrations in exceedance of 82ppb. Only site 1 had levels during the study period that exceeded the N A A Q O , this occurred for three peak hours at the beginning of the season in a period not considered to constitute an episode (see App. B.2). The time series shown in figure 4.9 further illustrates the decrease in diurnal variation with increased elevation that has been discussed previously. A t site 1 ozone concentrations are highest throughout the late afternoon (>50ppb for three days). A t this lowest site a complete ozone depletion to Oppb occurs in the early morning hours on all days but one; July 24, 2002 (JD 205) during the peak of the episode. Sites 3 and 4, at 600m and 1200m elevation respectively, both exhibit peak concentrations for the five days in exceedance of 50ppb. However, in the early morning minimum ozone values at site 3 are generally on the order of lOppb lower than at the higher elevation, site 4 having early morning minimums around 30 -20pbb. This further illustrates the decrease in diurnal variation, caused by a lack of ozone depletion, at higher elevations. The highest ozone concentrations at site 4 (1200m) occurred in the late afternoon during the July 2002 episode, with a secondary peak in the early morning when ozone at the lower elevation sites is at a minimum. This is unlike the 2001 episode when site 4 concentrations peaked around midnight However, it is well illustrated by figure 4.9 that nocturnally, ozone concentrations increase with elevation due to atmospheric stratification, whereas during the day vertical mixing distributes ozone more evenly within the boundary layer. Similarly, this general pattern was found during the August 2001 episode, and throughout both the 2001 and 2002 study seasons. The ozone episode described above also affected other parts of the L F V including municipalities between the study site and the city of Vancouver. Figure 4.10 gives a time series of hourly ozone averages for July 19-28, 2002 at Burnaby Mountain, Pitt Meadows, Maple 53 Ridge and Abbotsford, along with hourly average temperature at Abbotsford. Ozone concentrations in excess of 50ppb occurred at each of the sites during the peak of the episode (JD 204-206). Figure 4.10 - Regional hourly average ozone concentrations at Burnaby Mountain, Pitt Meadows, Maple Ridge and Abbottsford for the July 19-28, 2002 episode. Hourly average temperature at Abbottsford is also included. The suburban sites exhibit a diurnal pattern in ozone concentration closely related to temperature. Ozone precursors near the monitoring stations cause ozone to be produced at the times of most intense solar heating, and to be measured upon production. The Vancouver plume is not as apparent in this data as it was in 2001, by a time lag in maximum ozone concentration at sites further from the city. The Burnaby Mountain location, of a higher elevation than the other three stations, exhibited a secondary night-time peak in ozone concentration similar to that seen at site 4 for the same period. Also, Burnaby Mountain has peak concentrations lower than those of the other (lower elevation) suburban sites for the same period, similar to site 4's lower peak concentrations in 2002 than 2001. It is thought this 54 pattern may have been induced by more subsidence and a shallower boundary layer in 2002 uiat would concentrate pollutants nearer to the ground such that higher elevation monitors my be above the most heavily polluted air mass. However, temperature soundings are unavailable for either study season and it is difficult to determine boundary layer depth by other means. The July 2002 episode was dominated by synoptic events characteristic of above average ozone concentrations. Mean sea level (MSL) pressure maps showing the region from 40° - 60°N and 135° - 100°W are given in figure 4.11. Maps are given for July 19* (ID 220) figure 4.11a), 23 r d (JD 204) figure 4.11b) and 28* CD 209) figure 4.11c) 2002. The maps show the sub-tropical high pushing into British Columbia from the southwest and the development of a thermal trough along the west coast extending northward from the United States. This pattern is weaker than in 2001 - the isobars are more widely spaced. Both of these surface features are associated with above average ozone Concentrations (McKendry 1994), but the weakening of these features, relative to 2001, may account for the lower ozone concentrations than seen in the previously discussed episode. However, this thermal trough would still weaken the westerly surface winds, leading to some stagnation and thus allowing for ozone to build up in the western portion of the L F V during this period. These same conditions are responsible for the high daytime temperatures that were recorded, and other weather conditions favouring ozone formation and persistence (McKendry 1994). A t 500mb for the same area and time period (figure 4.12) there was no upper level ridge development even though it is a synoptic feature most commonly associated with ozone episodes (McKendry 1994). Instead, a split in flow induces weaker upper level pressure gradients. This is one of the few upper level flow scenarios when increased ozone concentrations occur in the absence of a ridge. In general, the highest concentrations of ozone 55 Figure 4.11 - One day means of sea level pressure in millibars (mb), lor July 19th, 23rd and 28th, 2(X)2. Source: NOAA -C IRES Climate Diagnostics Center, Boulder, Colorado, USA. From their website http://\unv.cdc.noa.gov/ 56 Figure 4.12 - One day means of.500 mh geopotential height in decametres, for July 19th, 23rd and 28th, 2001. Source: NOAA -C IRES Climate Diagnostics Center, Boulder, Colorado, USA. From their website http://www.cdc.noa.gov/ 57 occur when an upper level ridge is combined with the high surface pressure and thermal trough along the coast (McKendry 1994). Since the upper level ridge was not present, ozone concentrations during the July 2001 episode were lower than for other high ozone events such as the one in August 2001. It is also possible that due to a lack of hourly ozone data for much of the 2002 study period - especially August - that ozone episodes comparable to that witnessed in August of 2001 occurred, but were not recorded. During the period for which data is absent, there may have been a time when synoptic-scale conditions were dominated by an upper level ridge, combined with high surface pressure and thermal trough along the coast These conditions may have caused ozone concentrations to exceed the N A A Q O of 82ppb. However, after examining M W L A P ozone data for August it appears that the combination of conditions required for extreme ozone events did not occur even during the period of data loss. In August 2002 none of the suburban sites illustrated concentrations above and beyond 70ppb (except for one hour on the 28* of August when Abbottsford's ozone reached 79ppb), as they did during the August 2001 episode. 4.4 Summary of Altitudinal and Seasonal Variation In general, spatial ozone distribution is controlled by the stratification of the boundary layer, localised mountain flows, distance from source areas and other local features while temporal differences are due to a variability in meteorological parameters - both measured and unmeasured. The vertical variation in average ozone concentration with elevation is largely due to; localised mountain-valley flows, larger scale atmospheric stratification, and the weak diurnal variation experienced at higher elevations. 58 Mean ozone concentration tends to increase with elevation on south-facing slopes of the L FV . This pattern is not surprising as it has been found previously in a number of studies (Comrie 1990; Puxbaum et al. 1991; Aneja et al. 1994; Sandroni et al. 1994; Van Ooy and Carroll 1995; Zaveri et al. 1995; Banta et al. 1997; Mi l lan et al. 1997; Pisano et al. 1997; K lemm et al. 1998; McKendry et al. 1998; Brace and Peterson 1998; Cooper and Peterson 2000; McKendry and Lundgren 2000). It was found that on average site 4 (highest elevation) also represents the site of highest daily peak ozone concentrations, which is partially responsible for it having the highest seasonal mean ozone concentrations. The high mean ozone levels at the 1200m site are also the result of less nocturnal depletion at higher elevations. The relatively low levels of nitrous oxides at high elevations in rural areas hinder the chemical decay of ozone at night (Brace and Peterson 1998). Nocturnally (21:00 - 4:00 PST), ozone concentrations increase with elevation due to atmospheric stratification, local winds and a lack of nocturnal decay with altitude. The other trend responsible for the increase in seasonal mean ozone concentration with altitude, is the decrease in diurnal variation. A t high elevations, ozone levels do not decrease at night to the same extent as at lower sites. Bohm et al. (1991) found a strong correlation between high mean ozone concentrations and a lack of diurnal variation in ozone levels. Diurnal variation in hourly average ozone decreases with increased altitude in the LFV . This pattern is common for ozone in other regions of complex terrain (Puxbaum et al. 1991; Aneja et al. 1994; Sandroni et al. 1994; Van Ooy and Carroll 1995; Zaveri et al. 1995; Griinhage and Jager 1996; Brace and Peterson 1998). In this study of south-facing slopes of the LFV , local circulation is predominantly mountain-valley winds, and sea- or land- breezes. South-facing slopes are particularly 59 susceptible to high ozone levels in the summer due to radiation geometry that effects ozone formation within a plume, and strengthens the anabatic winds that transport pollutants to sub-alpine areas (Sandroni et al. 1994). Daytime anabatic winds allow ozone-rich air to be transported from low-lying valleys to areas of high elevation, where it remains aloft following the formation of a night-time inversion (Brace and Peterson 1998). Night-time upper-level subsidence and a temperature inversion at the surface create a residual layer aloft, that is characterised by elevated pollutant levels and remnants of the previous day's mixed layer (McKendry et al. 1997; McKendry and Lundgren 2000). The persistence of a residual layer throughout the night keeps pollutant concentrations elevated at higher elevations, such as site 4. If a site does not intercept the residual layer direcdy, local topographically produced katabatic winds can carry ozone rich air from aloft down slopes. This process allows increases in levels of ozone on higher elevation slopes during the night-time (Wanner et al. 1993; Zaveri et al. 1995; Brace and Peterson 1998) and may be responsible for the secondary peak seen in the early morning. The early morning is a time of peak radiative cooling, and hence the strongest katabatic winds. Alternatively, the secondary nocturnal maximum may due to the temporary turbulent coupling of the ozone-rich residual layer air to the surface often by the presence of a low-level jet (Corsmeier et al. 1997; Salmond and McKendry 2002). Conversely, low elevations are within the stable boundary layer at night where wet and dry surface deposition, and chemical decay, rapidly deplete ozone from the air (Banta et al. 1997; Glisten et al. 1998). This leads to a greater diurnal variation in ozone concentrations for the sites at lower elevations (McKendry and Lundgren 2000). The increase in Os levels for site 1 at 800h is consistent with residual layer down-mixing caused by the growing mixed-layer (Neu etal. 1994; McKendry etal. 1997; Gusten etal. 1998; McKendry and Lundgren 2000). These 60 day-old pollutants may contribute to the lower elevation's relatively high mean ozone levels. During the day, all sites show more uniform ozone distribution with elevation, due to boundary layer mixing (Zaveri et al. 1995; Pisano et al. 1997). Diurnal patterns of the three lower elevation sites show mean hourly ozone levels to peak in the late afternoon or early evening, decrease in the night-time and begin to rise again in the morning (=*8:00 PST). M in imum ozone concentrations occur in the early morning (400 -800h PST). The morning increase and night-time decline are consistent with previous findings of diurnal ozone variation (Fast and Zhong 1995; Banta et al. 1997; Pisano et al. 1997; Giisten et al. 1998; McKendry and Lundgren 2000). O n the other hand, the late afternoon peak is contrary to some studies and intuition, which found peak concentrations around, or shortly after, solar noon (Pisano et al. 1997) when ozone production rates are at tiieir highest However the study area, being downwind of the source region (Vancouver), will typically experience peaks occurring later in the day than areas closer to sources of ozone and precursors (Kelly et al. 1984; Edmonds and Basabe 1989; Sandroni et al. 1994; Van Ooy and Carroll 1995; Griinhage and Jager 1996; Krupa and Kickert 1997; Brace and Peterson 1998; Cooper and Peterson 2000). The average late afternoon peak in ozone concentration measured at all four sites, is due partially to the travel time it takes a plume formed in the urban environment to travel eastward into the rural areas of the LFV . The late afternoon peak in temperature occurring further east in the L F V also influences the afternoon peak in ozone. Although there is substantial decay of ozone, even during daytime hours, a plume may still be transported great distances, sometimes even acquiring or photochemically producing more Os as it travels (Comrie 1990). This may account for the higher concentrations at the further downwind M W L A P sites, for as Vancouver's urban plume travels eastward to Maple 61 Ridge it may pick up ozone and precursors from Burnaby, Abbotsford and Pitt Meadows. As distance from the urban source area increases, diurnal variation in ozone concentration may also decrease due to a lack of nitric oxide sources required for ozone scavenging (Bohm et al. 1991). The period of August 8i'-17'x, 2001 was marked by higher than average ozone levels. During this time sites 1, 2 and 3 show a similar pattern in diurnal variation, and again diurnal variation decreases with site elevation. During the episode all sites (except site 4) had peak ozone levels in the evening, and the lowest levels in the morning or early afternoon. This pattern also occurred at the four suburban M W L A P sites. Site 4, the site of highest altitude, showed the highest peak and the peak that occurred latest in the day with a very prominent late night maximum. Site 4 exceeded the N A A Q O for five hours during this episode, and its after midnight maximum in ozone concentration is consistent with other findings (Aneja etal. 1994; Zaveri etal. 1995). The August 2001 episode occurred in the presence of a surface high pushing northward, coupled with the development of a thermal trough along the west coast These were combined with an upper level ridge just east of 120°W, providing optimum conditions for the development of a high ozone episode (McKendry 1994). The episode in July of 2002 was characterised by weaker surface pressure gradients and the absence of an upper level ridge, leading to a less severe episode and the N A A Q O never being exceeded. Episodic ozone increases are important as they can lead to plant injury in sensitive species, as the frequency of peak exposures is more important than mean concentrations (Krupa and Kickert 1997). 62 5.0 W I T H I N C A N O P Y V A R I A T I O N I N O Z O N E A 10.5m tower measured vertical ozone variation using passive samplers at seven heights in a forest canopy. Each measurement height had two samplers for the purpose of replication. Relative humidity and temperature were measured at heights of 0.36m, 4.65m and 9.60m. W i n d speed and wind-direction were measured at the top of the tower (10.5m). Measurements were taken during five periods of two weeks in length and the result from this sampling follow. Recall that all meteorological data was lost during download for the period from August 8* - 27"' and therefore, will not be included in the following discussion. This chapter examines sunuriertirne vertical ozone profiles within the forest canopy and suggests possible processes responsible for the trends. 5.1 Passive Ozone Sampler Results Due to the nature of passive sampling, ozone is measured cumulatively. Ambient ozone reacts with the nitrite in the filter to form nitrate giving a cumulative ozone exposure. The nitrite to nitrate ratio attained through ion chromatography is used to find the total amount of ozone that the filter was exposed to over the two-week period in ppbh (see appendix A). This cumulative value is then divided by the number of hours in the sampling period to give an average ozone exposure, in ppb, used to create the vertical ozone profiles shown in figures 5.1 a-e. The aforementioned profiles represent ozone exposure within a small clearing in a low forest canopy (3-13m tall trees). Ozone increases with height in the forest canopy for all sampling intervals. For each of the five periods a strong power-law relationship was found between ozone exposure and height above the ground. Coefficients of determination range 63 a) b) 1000 800 1600 I 1400 200 d) 1000 800 ?600 JO i X 400 200 1000 800 8600 i »40O 200 July 9-23 Power (A) Power (B) B) y - 2E45x* m° -j/ . • r8-0.9815 yj A) y - 2E-06x6*"': i*-0.9718 .- j 6 8 . 10 J2 .  L4 16 .18, 20 22 24 Average Ozone Concentration (pp5, August 8-20 P J  1 A)y-5E-llx" , J M r» - 0.9464 W i im '// H /-f 1" 1 1 " l 1 1 1 B) y - 8E-06x61su r1 - 0.9741 10 12 14 16 18 20 22 24 Average Ozone Concentraion (ppb; 1000 800 0 600 400 Aug.20 - Sept. 3 X 200 B)y-33.358e 0'™' r1 - 0.8367 y / / s — / - * - ' - — J M f A)y-3E-09x8"* / IS - 0.9792 , 6 8 10 12 14 16 18 20 22 24 Average Ozone Concentration (ppb! e) September 3-17 A)y-0.0003xi4MS A r8 - 0.979 di Jj B)y-0.0245x&68i9 7jr r1 - 0.9674 W ) ! 10 12 14 16 18 20 22 24 Average Ozone Concentraion (ppb! 1000 - i 800 Sept. 17 - Oct. 1 Q 600 •it | 200 A)y-0.0013x™ 1 ! r 2 - 0.9664 B)y-0.0015xtM4! r8 - 0.9836 8 10 12 14 16 18 20 22 24 Average Ozone Concentraion (ppb! Figure 5.1 - Hourly average ozone concentration with height in a forest canopy for five 2-week periods. Data was collected as cumulative ozone (ppbh) for each period and converted to hourly averages (ppb). The five periods are a) July 9-23, b) August 8-20, c) August 20 -September 3, d) September 3-17, and e) September 17 - October 1. 64 between R2=0.9464 and R=0.9836, save the profile for side B from Aug.20-Sept.2 which has R2=0.8367. This relatively low coefficient of determination is due to the lowest sampler having an ozone exposure of Oppbh, which is considered to be an error in laboratory procedure. However, if this 0 value is removed there are too few data points to make any significant correlation. In addition, it was found that for those data points containing the error an exponential trend4ine fit better visually than a power-law one. Both the maximum and minimum ozone exposures in the profile decreased as the season progressed. This decrease in mean ozone concentration is due to decreases in air temperature and solar radiation from July to October, and is similar to the seasonal pattern exhibited by the average concentrations obtained through continuous monitoring (see App. B). The difference in relative concentrations between the top and bottom of the canopy, decreases towards the end of the season. The difference ranges from 9.3ppb for the first sampling period (July 9* - 23 r i, 2002) to 7.4ppb for the last of the five periods (Sept 17* - O c t 1", 2002). This progression towards a weaker vertical gradient in ozone concentration as the season advanced, is illustrated by an increase in the rate of change (slope of trend-line) as the season turned from summer to autumn (see figures 5.1 a-e). Conversely, the period from August 20 -September 2 (side A) had a more uniform within canopy ozone concentration with a difference of only 5.653ppb between the canopy top and bottom. This period was marked by a cooling trend compared with other sampling periods and similarly, both temperature and relative humidity also showed a weaker within canopy gradient for this period. 65 5.2 Meteorological Results Temperature was found to increase linearly with height in the canopy and to decrease over the study season (figure 5.2a). This decrease in temperature in the fall is responsible for the shift of ozone profiles to lower concentrations, discussed in the previous section. The a) 12 10 8 a 3,6 o X v-.5.8209x -60.311 y • 6.3045x - 7.5.48 y - 4.5451x - 77.373 I s-0.9481 i*-0.948 i*-0.9985 f y - 10.492x- 154.06 f \ 1^-0.9252 / / - f i I / • July 9-23 • Aug. 27 - Sept 3 • Sept. 3-17 • Sept. 17 - Oct. 1 1 1 1 • A • 10 12 14 16 Temperature fC) 18 20 22 b) 12 10 8 s a E 4 2 0 y~-0.9136x +71.668 i*-0.9948 y --1.3053x+ 122.15 r 0.9757 y--1.109x+102.3^ \ r-0.9713 \ • July 9-23 • Aug. 27 - Sept 3 A Sept 3-17 • Sept 17 - Oct. 1 \ y --3.5879x + 312.75\ \ \ ^ r 5-0.9681 f»*\A\ \ \ \\ \ 1 60 65 70 75 80 85 Relative Humidity (%) 90 95 100 Figure 5.2 - Wi th in canopy vertical profiles of biweekly mean temperature a) and relative humidity b). Measurements were taken at 0.36m, 4.65m, and 9.6m above the surface. 66 period from August 27* - September 3 r d, 2002 shows the weakest gradient in temperature with height in the canopy (m ~ 10.5). This same period showed the weakest vertical gradient in ozone. However, meteorological data is missing for August 20* - 27*, 2002 so the mean temperature is based on half the number of hours, (n = 168) of the ozone data (n = 336). Relative humidity (figure 5.2b) was also relatively uniform with height in the canopy for this period (m * -3.6), and therefore it is believed that this time was marked by enhanced vertical mixing thereby leading to weaker gradients in ozone, temperature and relative humidity. Relative humidity decreased linearly with height in the forest canopy and showed a general increase over the season towards autumn (figure 5.2b). The high relative humidity towards the end of the study period, combined with the lower temperatures, would contribute to the decline in ozone concentrations, since ozone production is the highest in times of warm temperatures and low relative humidity (Edmonds and Basabe 1989; Aneja etal. 1994; Giisten etal. 1998). Not surprisingly, it was found that temperature and relative humidity had a negative correlation in the forest canopy (figures 5.3a and 5.3b). Temperatures decreased towards the end of the monitoring period and the vertical trends in both temperature and relative humidity discussed in the previous section can be made out slightly from the time series data represented in the figures. The period from August 27 - September 3 rd, 2002, when within canopy gradients were the weakest, shows a cooUng trend (figure 5.3a). It appears that the temperatures were very warm prior to and at the beginning of this period approaching 30 ° C. The high temperatures may have lead to convective mixing and uniform distribution of heat, ozone and water vapour. Cooler temperatures may have lowered source emissions keeping ozone lower at the canopy top and therefore closer to surface concentrations, reducing the 67 a) 0.36m ? P ? ^ s n i ; s s s s s a 3 8 ^ _ _ r H r-< r-H — , ^ H ^ H r H r - t r H , - l ^ H ^ - l « « Julian Day b) 0.36m Julian Day Figure 5.3 - Seasonal trends in hourly average temperature and relative humidity for a) June 21-23 and b) August 27 - October 1. 68 a) W i n d Direction -Windspeed o •8 P P f2 ^ E3 83 S S S S g S S 8 3 8 Julian Day b) 360 300 | 240 •a 180 120 60 0 W i nd Direction -Windspeed J — , ^ . __— . 1 • • * • * • • • • • » • • • * • A. . * • • • • • • • k 1 •* , . • # t • • * * * • J - J I ' " i » i t 1 » - • • i i i * : JfcjLli llk.-liy i LA KJLti • • * : * w r • * • \ ' • * ., -Julian Day 4.1 & Figure 5.4 - Vectors of windspeed and direction measured on the tower at 10.5m height in canopy for a) June 21 - July 23 and b) August 27 - October 1. 69 vertical ozone gradient in the absence of convection. A cooler moister air mass moving into the area would have temperatures and relative humidities closer to that of the forest floor and may have led to the weaker gradients in temperature and water vapour. Figure 5.4a and 5.4b show horizontal mean wind at 10.5m for the study period from July 9* - October 1", 2002 Julian days 171-274). W i n d direction at the top of the tower was J found to be extremely variable throughout the study period. However, the wind was light throughout the season, within mean hourly winds never exceeding 6m/s. Although the variable wind would suggest turbulence, the frequendy calm wind would mean a lack of turbulent mixing by dynamic instability. Static instability discussed above is more likely responsible for ozone down-mixing to the surface. This is further illustrated by the period from August 27* -September 3 rd, 20020D 239-245) when the distribution of various elements was the most uniform with height. Figure 5.4b shows that during this period winds were less than 2.2m/s, however the highly variable wind direction suggests turbulence. The low wind-speeds imply little windshear, hence this turbulence is probably related to static rather than dynamic instability, from the release of latent heat during this period of cooling. 5.3 Factors Affecting within-Canopy Ozone Profiles There are various factors believed to contribute to the strong power-law relationship between ozone exposure and height in the forest canopy. The most important of these factors are illustrated schematically in figure 5.5. The following will be a discussion of these within canopy processes and will refer to each process in the figure by number. Firstly, a combination of mean horizontal wind and surface roughness leads to the turbulent down-mixing of ozone (and other atmospheric constituents) into the forest canopy 70 ®. The stronger the wind and the more variable the upper canopy surface layer the more turbulent the flow will be (Stull 2000), and hence the more down-mixing. In the middle of the day when wind-speeds are the highest, turbulent transfer will penetrate more deeply into the forest canopy (Oke 1987) and ozone will reach the surface. The warming of the lower canopy during periods of high insolation may lead to increased convective transport to lower surfaces (Wesely et al. 1982). This waiming is greater for more open canopies with a low L A I (leaf area index) and may lead to enough within canopy turbulence to create down-mixing (Novak et al. 2000; Lamaud et al. 2002). The conifer canopy is relatively more open than deciduous canopies of previous studies, and this warming process, coupled with daytime winds, would create sufficient turbulence required for the down-mixing of ozone to the surface. Surface-atmosphere instability from the release of latent, rather than sensible heat, may also contribute to vertical mixing as discussed previously. Once at the surface, or forest floor, deposition and chemical destruction act to deplete ozone ©. Dry deposition occurs on non-stomatal surfaces including soil, trunks, and leaf cuticles since ozone is not water soluble (Lamaud et al. 2002). Also, chemical reactions with nitrogen oxides emitted from the soil lead to chemical destruction at the surface. Since no measurements of N O or NO* were made and dry deposition rates were left unmeasured, it is impossible to accept or reject this hypothesis. However, these depositional processes are thought to contribute significandy to the patterns in ozone concentration that were observed. Since ozone is not water soluble, it has in the past been thought to be depleted near the surface via processes of 'dry' deposition. The 'wet' deposition often experienced is explained by the occurrence of chemical species present in water that react with ozone to destroy it (Lamaud et al. 2002), and the moist humid surface environment may provide optimum conditions for wet 71 deposition. Deposition velocities are the highest under conditions of low light, such as would exist in the forest understorey and at night (Kerstiens and Lendzian 1989), thereby contributing to ozone depletion at the surface. Low or young trees and other mid-canopy vegetation may take up some ozone direcdy through leaf stomates (D. A t the tower location, there were many young Alders (Alnus rubra), that may have acted as a sink for ozone. Uptake by vegetation may be an important factor leading to the power-law relationship, especially if understory uptake is high @. Lamaud er al. (2002) found ozone uptake by under-story plants to account for a significant portion of ozone deposition in a pine stand. The uptake rates of ozone by vegetation my account for much higher ozone depletion than previously accounted for by dry deposition alone. Under-story contributions to ozone fluxes have been shown to be greater than the under-story contributions to either fluxes of sensible heat or water vapour (Krupa and Kickert 1997). Additionally, Lamaud etal. (2002) found the presence of moisture on the under-story surface (associated with dew) to enhance ozone uptake. The increase in relative humidity towards the forest floor and the periodic visible wetness of vegetation, suggests that uptake of Os by quick growing under-story shrubs may be an important factor in ozone reduction at the surface. Relative humidity was often 100% (see figure 5.3b) especially near the surface (0.36m) and Lamaud er al. (2002) found surface conductivity to ozone to increase sharply as the relative humidity approached 100%. It has not yet been determined why moisture increases uptake, however stornatal conductance, and hence the uptake of gases by a leaf, is positively correlated with moisture availability. It is therefore not surprising that uptake would increase near the canopy floor where neither moisture-stress nor heat-stress contribute to stornatal closure that would reduce gaseous uptake. 72 Figure 5.5 - A schematic diagram oi factors that influence the vertical distribution ofozone within a forest canopy. Numbers correspond to different processes and arc discussed within the text. If the wind is calm and/or the forest very homogenous (i.e. trees of same age and species - a plantation) there will be minimal dynamic instability to promote down-mixing. Similarly, if there is a lack of convection, static stability may prevail and down-mixing ® may not occur. A t night when winds are very light, turbulent transfer does not penetrate as deeply into the canopy (Oke 1987). Ozone may only be mixed to a very shallow depth in the forest 73 and not reach the surface (D. This mosdy nocturnal process leads to the power-law relationship overnight, when uptake by vegetation is minimal. In the night-time when uptake by plants is low, dry deposition, fed by dynamic instability, is said to be responsible for depletion at the surface (Lamaud et al. 2002). However, since wind speeds remain low nocturnally, sufficient wind-shear may not be present to create the instability required for dry deposition at the surface, especially in the absence of convection to promote mixing. In conclusion, it is believed that cumulative mean surface declines in ozone concentration are due to daytime convective mixing leading to surface deposition (D & © and uptake by vegetation ® & (D. A t night the power-law relationship persists due to a lack of ozone mixing to the surface. The power-law relationship may be further strengthened by the ability of some species to take up ozone at night, thereby affecting nocturnal vertical profiles as well as daytime profiles (Emberson et al. 2000). Surface wetness (estimated by relative humidity) may act as a stronger sink for ozone deposition than the stornatal cavities (Lamaud et al. 2002). When plant stomates are closed, ozone is eidier destroyed chemically at the leaf surface or the leaf cuticle, but ozone may also enter the leaf via diffusion - controlled by intercellular ozone concentration (Krupa and Kickert 1997). Ozone deposition velocity may exceed ozone permeance due to ozone decomposition at the cuticle surface before diffusion takes place (Kerstiens and Lendzian, 1989). Deposition, uptake and a lack of nocturnal mixing are all partially responsible for the vertical witfiin canopy ozone profiles in Figures 5.1 a-d. DespitE the strong vertical gradients of ozone within the canopy, growth reductions may not be canopy-level-specific because concentrations remain high at the canopy crown where uptake and carbon assimilation are the highest for forest trees (Ollinger et al. 1997). 74 6.0 I M P A C T S O N F O R E S T V E G E T A T I O N Little, if any, information is available on the effects of tropospheric ozone on the native forests plants of western Canada. Vegetation impacts in this study were therefore, assessed based on European and U.S. standards and studies. A t the end of July 2002, all vegetation at the four continuous sampling sites was examined for any discoloration to foliage that may be indicative of ozone injury. Individual broad-leaved shrubs were chosen for monitoring - with the exception of a Douglas-fir sapling at site 2. Plants displaying foliar symptoms, indistinguishable from those caused by ozone, were chosen for biweekly observation throughout the study period. Results include injury scores using the Horsfall-Barratt scale (HB-scale) and digital photographs of symptoms. The European (AOT40) and American (SUM06) indices, developed to protect vegetation from ozone injury, were also calculated for both 2001 and 2002 and include 24h data for each day of the four month study period. These indices examine exposure to vegetation through cumulative dose analysis and will be discussed in the following section with reference to forest injury in the LFV . 6.1 Seasonal Development of Cumulative Ozone There is some disagreement within both the scientific and political communities regarding the threshold ozone concentration above which damage to natural vegetation occurs. This uncertainty is reflected in the differing air quality indices developed to protect vegetation, that are used in this study. Two metrics are described in the following section and include the European (AOT40) and American (SUM06) standards. 75 6.1.1 AOT40 Cumulative ozone exposure above 40ppb (AOT40) for the summer of 2001 is shown in Figure 6.1. There is a marked increase in cumulative ozone exposure with elevation, similar to the increase in average ozone concentration with elevation discussed in section 4.1. The European A O T 4 0 index, established to protect vegetation, was calculated for the 2001 and 2002 study seasons. The August 8 - 17 episode is clearly apparent (Julian days 220 - 229) and contributes to the majority of cumulative exposure above 40ppb. Site 4 experiences the highest accumulated exposure with an A O T 4 0 index of 5209ppb.h. In general this value may be considered low, since over a 6 month period (usually Apr i l - September) the cumulative A O T 4 0 Level I Critical level for forest trees is set at 10 OOOppb.h (lOppm.h) (Karenlampi and Skarby 1996). Above this critical level, injury vegetation occurs (Ghosh et al. 1998). However, others including Karlsson etal. (2002) have found that A O T 4 0 values of less than the lOppm.h may pose a threat to vegetation. VanderHeyden et al. (2001) have found ozone injury to occur at A O T 4 0 values below 5000ppb.h. The A O T 4 0 index was designed to protect agricultural crops and was formulated through laboratory studies, in the natural environment a lower threshold may be required to cause direct injury to vegetation. In addition, A O T 4 0 was calculated for four months of cumulative ozone. Hypothetically, if six months of data were included and there were no gaps in data, it is possible, though unlikely, that an A O T 4 0 of 10000 ppbh would have been exceeded at site 4. The other three sites had low cumulative hourly ozone exposures over 40ppb of 733ppbh, 991ppbh, and 1414ppb for sites 1, 2, and 3 respectively. A t lower elevations there appears to be a very low risk of injury to forests based o n A O T 4 0 . 76 6000 5000 4000 o 3000 2000 1000 •Site 1 •Site 2 •Site 3 •Site 4 b 2 2 2 g § S S S § S ^ S S S ^ Julian Day Figure 6.1 - Cumulative A O T 4 0 at all sites for the 2001 study season 2001 as can be expected by the yearly differences in ozone and meteorology discussed in the previous section. Accumulated ozone exposures above 40ppb are shown in Figure 6.2 for 2002. Each of the three sites' A O T 4 0 cumulative exposures are below lOOOppbh - a tenth of the yearly threshold required to induce injury. Sites 1, 3 and 4 have an accumulated A O T 4 0 of 935ppbh, 704ppbh, and 915ppbh, respectively. The low A O T 4 0 values are in part due to the meteorology discussed previously ttiat lead to the general yearly differences in ozone concentrations. Additionally, 3 - 6 weeks of data are missing for each of the three sites in 2002 (see App. B.2), making the A O T 4 0 scores representative of 2 or 3 months rather than 4. Some important episodes, similar to the August 8 -17, 2001 episode that contributed largely to AOT40 , may have been missed. However, the smaller increases constituting less severe episodes from July 8 - 14, 2002 (189-195) and July 19 - 27, 2002 (200-208), appear the most important times of ozone accumulation recorded over the 2002 study season (see figure 6.2). 77 6000 5000 4000 S 3000 I i 2000 1000 0 4w •Site 1 •Site 3 •Site 4 Mill IIP Ml 11 IT IITITnTTM Iffiri N i I  I I  II IM i n 11 IM i F2 p I GO Oi Oi O •—I »—« CO — . — < . — . OI CM C«l OS OA <M Ju l i an D a y to c<i co I-I W ) HQ <0 l '- C 3 o a o s o > co o i Figure 6.2 - Cumulative A O T 4 0 at all site for the 2002 study season Fuhrer and Acherman (1997) found that injury symptoms occur just as frequendy at concentrations above 30ppb as they do above 40ppb. Although correlation coefficients between concentration and injury were comparable at 30ppb, this concentration is low enough to be considered a 'natural' level (i.e. without anthropogenic influence), and therefore, the A O T index is based on concentration in exceedance of 40ppb. 6.1.2 SUM06 The increase in cumulative ozone exposure with elevation is also illustrated through calculation of the American SUM06 metric. Cumulative ozone levels above 60ppb for the summer of 2001 are shown in figure 6.3. The August 8 - 1 7 episode is again apparent (Julian days 220 - 229) and contributes to the majority of ozone concentrations over 60ppb. Site 4 experiences the highest accumulated exposure with a SUM06 index of 5909ppbh. Again, this 78 Jul ian Day Figure 6.3 - Cumulative SUM06 of all sites for the 2001 study season index was designed to protect agricultural crops and formulated through laboratory studies, it is therefore not optimum for forest application. In addition, rather than a single threshold approach, such as that used with AOT40 , SUM06 bases exceedance thresholds on the percent of a crop that will be protected from a certain percentage of damage. For instance, a SUM06 of 26.4ppbh protects 50% of crops from experiencing more than 10% damage (Krupa and Kickert 1997). It is therefore difficult to say at which critical value of SUM06 forest plants may incur visible injury. Neither AOT40 , nor SUM06, take into account factors of atmospheric resistance, time of occurrence, humidity, soil moisture, stornatal conductance, and other key factors (besides hourly averages Os concentration) that strongly affect uptake rates and hence, injury (Krupa and Kickert 1997). The otiier three sites had low cumulative ozone exposures above 60ppb of l l l l p p b h , 710ppbh, and 327ppbh for sites 1, 2, and 3 respectively. A t lower elevations there appears to be a very low risk of injury to forests based on either index. 79 6000 ~i 5000 I 4000 1 1 3000 1 2000 1000 -0 •Site 1 •Site 3 • Site 4 —• —. —. —. CM C1 C-\ 0\ Ol OI CM C-» CM C-1 CM CM Jul ian Day Figure 6.4 - Cumulative SUM06 of all sites for the 2002 study season The SUM06 values calculated for the 2002 study season are much lower than those in 2001 as can be expected by the yearly differences discussed in the previous section. Accumulated hourly ozone exposures above 60ppb are shown in Figure 6.4 for 2002. Each of the four sites' SUM06 cumulative exposures are below 800ppbh. Cumulative SUM06 values are 734ppbh, 191ppbh, and 653ppbh, for sites 1, 3 and 4 respectively. The low SUM06 values are in part due to the meteorology discussed previously that lead to the general yearly differences in ozone concentrations. Also, 3 - 6 weeks of data are missing for each of the three sites in 2002, meaning dial the SUM06 scores are representative of 2 or 3 months rather than 4. There may have been an episode, similar to that from August 8 -17, 2001 mat would have contributed largely to SUM06 widi hourly ozone levels above 60ppb for extended periods. However, the smaller increases in ozone concentration from July 8 - 14 (189-195) and July 19 - 27 (200-208) appear the most important times of ozone accumulation recorded over the 2002 study season. 80 6.1.3 Discussion of Indices Although the index thresholds were not exceeded in either year, symptoms of ozone injury have been found to occur at exposures below the A O T 4 0 threshold (Innes and Skelly 1996; Bergmann et al. 1999; VanderHeyden et al. 2001; Karlsson etal. 2002; Novak et al. 2003). It is important to note that the A O T 4 0 threshold of lOppm.h was developed based on significant 10% reductions in annual biomass accumulation (Karenlampi and Skarby 1996) rather on the occurrence of foliar injury. Therefore, foliar injury to forest plants may occur at levels diat do not exceed this threshold. Bergmann et al. (1999) suggest that instead of being based on growth reductions, that exceedances of critical levels should be defined as concentrations at which any change in the frequency of a species or genotype occurs. This would require that critical levels be species' and or plant community specific. A documented problem of both the European and U.S. indices is the overestimation of the importance of ozone in the late afternoon/early evening when plant uptake rates are lowered (Krupa and Kickert 1997). This would be especially problematic when using 24-hour data to calculate indices, and for areas some distance from the urban source, where peaks occur later in the day, such as the L F V study areas utilised here. However, many plant species can have nighttime stomatal conductance resulting in ozone uptake, even though most uptake occurs during daylight hours, (Musselman and Minnick 2000). Nocturnal uptake is a function of a number variables, including species, region (e.g., alpine, grassland, deciduous forest, etc.), season, and elevation. In addition, many areas, particularly those in mountainous regions, have nocturnally high ozone exposures and plants may be more sensitive to ozone at night due to lowered defences (Musselman and Minnick 2000). The enhanced night-time exposures at high 81 elevations (1200m), including a near midnight peak, may prove important in assessing cumulative exposure to sub-alpine vegetation. 6.2 Foliar Symptoms and Injury Development Assessment of native shrubs for injury symptoms began on August 8*, 2002. Foliar symptoms occurring at each of the field sites as well as the arboretum were examined. Symptoms were initially, an upper-leaf interveinal discolouration or stipple that was red or purple in hue. A l l individuals within the immediate vicinity of each of the sites, that displayed symptoms identical to those of ozone injury, were included in the study. The one Conifer (Psuedotsuga menziesii ssp. menziesii at site 2) that was monitored displayed a chlorotic banding on most needles, a symptom indicative on ozone injury to coniferous tree species (Langebartels et al. 1998; Brace et al. 1999). The initial symptoms of each monitored individual are given in Table 6.1. Those described are somewhat 'classic' symptoms of ozone injury (Ghosh et al. 1998; Brace et al. 1999; Mil ler and Arbaugh 2000; Sanz and Mil lan 2000; Innes et al. 2001; VanderHeyden et al. 2001; Novak et al. 2003) and tended to increase over time as the cumulative ozone dose increased. Appendix C . l contains photographs taken of each of the monitored individuals on each of the biweekly assessment days. Images are of probable ozone injury, however environmentally specific and species' specific symptoms have not yet been verified. The categorical nature of the injury scores makes it difficult for usual time series methods to be employed in analysis (Ghosh etal. 1998). In addition each of the time series in this study had only five or six data points (the number of times an individual was assessed for injury). Due to these restraints data will be presented as HB-scores received by each individual 82 Arb. Spirca douglasi Upper surface purpling - not on veins r Vaccmium parvifohum \ * 1 Reddening on upper surface but not veins l < , 1 Rubus parvifolius I lght purpling on leaf maiguis Vaccmium parvifohum >t t Reddening on upper surface1 but not. veins ^  v ' 2 Psuedotsuga mcnzietu Chlorosis and chlorohc banding on nearly all needles ' * A }\Gaulthena shallon', I ^*, ' Redd emng/sdpple on upper surface but not veins k Vacomu/n parvifohum Reddening on upper surface but not veins • ' Acer circmauim < Upper surface red stipple - aieas furthest from veins 3 Gaulthena shallon Reddening/stipple on upper surface but not veins ^ ^Vaccmium parvifohum \ ' - ^'Reddening on upper, surface but not veins, \ , ' Vaccmium membraccum Reddening on leaf and vein margins (light supple) " \ t Ace i gldbrum ^ 1 '-*~\ ignf reddening;on upper ^ surface (bmshecl)^ Comus canadensis Deep reddening between veins, more pronounced towards Up '4 > Vaccmium membrae'eum \ Reddenmg on leaf mai gins and-between veins on upper rf t ^ 1 v \ ^ ' ^ surface, some sUpple ^ " y Vaccmium ovahfolium Reddening of leaf margins on upper surface, purplmg of upper surface Table 6.1 - Brief desenpuons of lniual foliar injury symptoms experienced by each individual monitored throughout the 2002 field season. for each monitoring day over the summer of 2002. The assessment of plant injury based on injury scores alone is further complicated by a reduction in the score that may be due to a progression of injury. Such a progression or worsening of injury symptoms that may reduce injury scores include newly damaged leaves with a lower percent of leaf area damaged, or the loss of injured leaves (Ghosh et ai. 1998; Novak et ai. 2003). Appendix C.2 describes the severity of potential ozone injury symptoms found on individuals in the summers of 2002. Individuals are listed by site location - Arboretum, site 1 (200m), site 2 (400m), site 3 (600m) and site 4 (1200m), and by species' name. The date of qualitative data collection is given, and a corresponding visual estimate of percent leaf area 83 (LA) injured and percent of total plant foliage (TP) affected, is provided. These two estimates are then multiplied to give a Horsfall-Barratt (HB) injury score. Unfortunately, much of the period during which vegetation was examined for injury lacks any hourly ozone data (see App. B.2). Therefore, the following discussion of symptom progression will include trends in cumulative ozone concentrations for the sites discussed earlier in this chapter, and elevational trends discussed in Chapter 4.0. Figure 6.5 illustrates this progression of injury symptoms over time using H B scores of injury severity. Seasonal and elevational patterns of increasing symptom development were found to occur on the monitored plants. Both of the individuals monitored at the arboretum (figure 6.5a) show an increase in H B injury score throughout the period. Although ozone concentrations were not monitored at the arboretum in 2002, it constituted site 1 in 2001 - a location with high daily maximum concentrations and the second highest mean concentrations for the season (see Table 4.1). The spikes in concentration that occur at this site may be enough to cause acute injury in some species, namely 5. douglasi and V. parvifolium whose symptoms worsened over the season which is indicative of ozone injury which is cumulative. A t site 1 in 2002, (figure 6.5b), the R. parvifolium that was monitored showed a progressive development of minor injury symptoms (HB>60) displayed as a slight reddening on leaf margins. The V. parvifolium at site 1 was deemed not to show symptoms indicative of ozone injury at the beginning of the season - discolouration was blotchy and affected veins. However, the low H B score (HB=60) given for the final assessment on September 19* was much more ozone-like, and although is was very mild it consisted of inter-veinal light reddening on leaf margins. The frequency of injury symptoms may have been lower at site 1 due to low ozone accumulation and low seasonal mean concentrations. However, this site experienced 84 a) 1.) 10000 8000 6000 4000 2000 e) Arboretum 10000, Site 1 (200m) 8000 6000 4000 2000 i Rubus psavifolius I Vaccmium parvifohum 8/8 8/20 ft/3 9/17 c) d) 10000 8000 6000 4000 2000 Site 2 (400m) 0 4— • Psuedotsuga menziesii • Vaccmium parvifohum • Gaultheria shaUon • Acer ckcinatum 10000 8000 6000 4000 2000 • Vaccmium membraceum • Vaccmium parvifohum 1 m Gaultheria shallon QAcer glabrum • Comus canadensis • V. parvifohum 2 Site 3 (600m) 10000 8000 6000 • Vaccmium membraceum • V. membraceum 2 • V. membraceum 3 • Vaccmium ovahfohum • V. ovahfohum 2 • V. ovahfohum 3 Site 4 (1200m) 8/27 9/10 9/24 Date Figure 6.5 - Horsfall-Barratt injury scores recorded for each of the monitored plants at the arboretum a), site 1 b), site 2 c), site 3 d) and site 4 e) for the 2002 study season. 85 the highest maximum ozone concentrations in 2002 (96ppb) and intermittent high ozone exposures may cause the infrequent and mild injury occurring on site 1 plants. Site 2 had no ozone data for 2002 however recall that in 2001 site 2, as well as site 3, showed mid-range means, mid-range maximums and intermediate diurnal variation. This absence of both high peak concentrations and consistendy high concentrations may account for the inconsistent patterns in injury development that occurred at both of these sites (see figures 6.5 c) and d)). Without ozone 'spikes' (>70ppb) acute injury can not occur, unless mid-range concentrations (>40ppb) persist causing chronic exposure. Despite these patterns of exposure, symptoms indicative of ozone injury were found at both sites 2 and 3. Psuedotsuga menziesii spp. menziesii, the only conifer examined for injury in the study was located at site 2 (figure 6.5c). The chlorotic banding expressed by this individual is consistent with conifer ozone injury symptoms (Brace and Peterson 1999; Mil ler and Arbaugh 2000). The H B score for this Douglas-fir was high and went from 5000 to 7500 over the study period. Th is symptom worsening may have been due to an increase in cumulative exposure towards the end of August, but due to a lack of data this hypothesis can not be tested. The Vaccinium parvifolium at site 2 experienced a decrease in H B injury score on August 20 due to a decrease in the average leaf area injured arid a decline in the percent of injured leaves, which may have occurred due to the development of new uninjured leaves. This decrease in injury was followed by an increase in injury in time for the next monitoring period and eventual premature senescence by September 17*, 2002 (see image App. C. l ) . This plant had lost most foliage by September 9*, 2002. The Gaultheria shallon monitored for injury at site 2 (figure 6.5c) shows decreases in H B score on two occasions throughout the monitoring period. The lowered injury score may 86 be due to a loss of injured leaves, or the development of new uninjured leaves. However the decrease in symptoms may mean the symptoms do not represent ozone injury, but instead another stressor for which injury is temporary and not cumulative. There was a late season increase in injury symptom severity for Acer circinatum at site 2. By September 17*, 2002 this individual possessed a H B injury score of 7500 (see App C.2) after having a H B of 5000 for the rest of the monitoring periods. Many previously uninjured leaves on this plant displayed injury symptoms on the last monitoring day. This was following the same period after which the foliage of V. parvifolium at this site withered and died. The Vaccinium membraceum at site 3 (figure 6.5d) displayed worsening ozone-like symptoms for the first two monitoring periods and then suffered from leaf abscission (see App. C.2), half way through the monitoring period. Consequently there is no H B score for this individual after August 20*, 2002. There were two different V. parvifolium plants displaying ozone-like injury at site 3. Both showed a general increase in symptoms through the season with the exception V. parvifolium 1 on the 20* of August, 2002, when there was new injury found on leaves not previously symptomatic. This lowered the average L A showing injury, and even though it increased the T P showing injury, the H B score was lowered (see App.C.2.) V. parvifolium 2 showed a decrease in H B injury score on September 3 r d, 2002 for the same reason and due to a partial loss of injured leaves such that the T P percent also decreased. Site 3's G. shallon suffered from minor ozone-like symptoms that seemed to fluctuate in extent and severity. FoUovving seasonal monitoring and an examination of symptoms, it was determined that this plant's symptoms were most likely not ozone-induced, but instead were indicative of another stressor. The most probable of which is a fungal infection common to this species (Humphreys 2001 pers comm.). Conversely, Acer glabrum symptoms increased 87 throughout the season and remained identical to those of ozone injury. There was some recovery during the period between August 8 - 20, 2002 when it appeared as though new leaves had developed and some injured ones shed. Cornus canadensis, like its Vaccinium spp. neighbours, suffered from leaf loss due to premature senescence, but recovered completely and gained new leaves that never displayed injury symptoms (see App. C. l ) . There was no increase in injury preceding this recovery, but the symptoms were indicative of ozone damage including the leaf-shed. A t site 4, only two different species were monitored due to the decrease in shrub and tree species diversity found with elevation. However, three individuals of both V. membraceum and V. parvifolium were included in the assessment of injury. V. membraceum displayed a general increase in injury over the season with a marked recovery period between August 13* and 27* when either new leaves developed or injured leaves were shed. In contrast V. ovalifolium showed mild symptom development without recovery throughout the monitoring period. Some foliage of V. ovalifolium 3 was very discoloured showing severe interveinal reddening on September 10* and 24* (see App. C. l ) and may be an example of severe injury leading to early senescence. Foliar injury symptoms generally progressed or changed throughout the growing season. According to VanderHeyden etal. (2001) and Mil ler etal. (1983) injury shows a typical pattern of progression. Ozone injury ordinarily begins as chlorotic mottle, then becomes an upper-leaf-surface stipple, followed by a general discolouration (necrosis), and lastly premature leaf senescence or abscission. It is possible the chlorotic spotting and or stipple had already occurred on the individuals prior to the first week in August, since by that time most expressed a general interveinal adaxial discolouration (see table 6.1). A t least two of the individuals 88 displayed symptoms that progressed to abscission (leaf mortality) and or early senescence (leaf loss). The premature shedding of foliage or early leaf senescence can cause the extent and severity of foliar symptoms to decrease due to the loss of symptomatic leaves (Ghosh et al. 1998; McLaughlin 1998). This occurred in the 2002 field study for a few individuals including Vaccinium parvifohum at site 2 and V. membraceum at site 3, which experienced premature senescence in early September. Upon return to the sites for injury assessment, the individuals appeared lifeless, in contrast to some cases where new uninjured leaves may replace injured foliage (Ghosh et al. 1998). Foliage replacement occurred for site 3's Cornus canadensis decreasing the injury score as time progressed. In both cases of the loss of injured leaves leads to an H B score that may decrease as the cumulative ozone (AOT40 or SUM06) increases, meaning that injury may no longer be thought of as cumulative. Although mid-season leaf shed is not common under ambient conditions, accelerated senescence towards the end of the growing season may occur (Ghosh et al. 1998). This study was conducted in late summer so that it would be early enough to miss natural leaf senescence but late enough for sufficient ozone to have been accumulated in foliar tissue. It is therefore possible that the premature senescence and foliar discolouration displayed by shrubs at the monitoring sites, was induced by chronic or acute ozone exposure. Acute exposure would occur at lower elevations (arboretum or site 1) where ozone is depleted at night and diurnal variation causes high daytime peak concentrations, especially during episodes. Symptoms of acute injury include chlorosis (chlorophyll leaching) and cellular pigmentation (stipple) (Krupa and Kickert 1997; Paakkonen et al. 1998; Brace et al. 1999). Acute injury causes death to individual cells leading to stipple. Chronic exposure would 89 occur at higher elevations (site 4) where ozone concentrations reside throughout both the day and night, and are the highest in terms of seasonal means. Injury from chronic exposure occurs gradually and is characterised by such symptoms as necrosis and premature leaf senescence (Brace et al. 1999). These symptoms occurred at the study sites, but due to the preliminary nature of the vegetation survey it is difficult to diagnose symptom causality as either chronic or acute. Chronic exposure has been suggested to be the greatest concern in terms of forest health due to affects on growth, whereas acute injury is of less concern due to symptom recovery during times of low ozone levels (Krupa and Manning 1988). However, others have suggested that acute exposure, expressed as high hourly averages, may be more of a threat to vegetation (Lefohn and Runeckles 1987; Sanz and Mil lan 2000). Therefore, the relative threat of injury due to chronic or acute exposure is deemed to be both environmentally and species specific, while being dependent on everything from a plant's natural defences and recovery time frequency to moisture and wind regimes. Plant recovery from ozone injury occurs late in the season, usually near the beginning of September. This is when ozone levels began to decrease due to seasonal changes such as cooler temperatures, higher humidity and lower intensities of solar radiation. This decrease towards September is illustrated by the shift of within canopy ozone to lower concentrations that occurred as the season progressed (see figure 5.1). Lowered ozone exposure allows vegetation to recover from injury incurred during periods of high ozone concentration such as episode's. The shedding of damaged foliage and the growth of new may increase the plants photosynthetic capacity. Therefore, vegetation growing in areas of high ozone concentrations where severe injury leads to leaf re-growth may have higher rates of photosynthesis than those 90 located where concentrations are too low to lead to complete senescence (Beyers et ai. 1992). Repair of plant tissue during low ozone periods is an important plant survival mechanism in affected areas (Kickert and Krupa 1991; Mil ler and Arbaugh 2000). 6.3 Discussion of Injury to Vegetation Plants take up ozone direcdy through stomates, similar to the uptake of other gases such a carbon dioxide and water vapour. Uptake rates of ozone by vegetation is therefore, strongly controlled by atmospheric and stomatal resistances, which are neglected by exposure metrics such as A O T 4 0 and SUM06 (Krupa and Kickert 1997; Emberson et ai. 2000). However, these metrics are simple, easy to enforce and are used widely in policy initiatives. Although ozone concentrations in the L F V did not exceed index thresholds in either 2001 or 2002, the levels may still have been high enough to cause injury. Ghosh et ai. (1998) found that even in charcoal air filtered chambers containing 50% ambient ozone, (24-hour A O T 4 0 values of 2692 - 4735ppb.h) plant species still displayed foliar injury symptoms. In these low ozone accumulation instances symptoms were less severe and had an onset later in the season, but still developed. Even when visible injury is absent damage to plants may still be occurring (Wang and Bormann 1986). Bergmann et al. (1999) found visible symptoms to occur on only 50% of the 118 individuals they tested. The effects of ozone on plants may occur at the cell, organ, whole-plant or population level and everything from nutrient and moisture availability, to temperature and wind-speed affect the uptake of ozone by plants (Sandermann et al. 1997). These factors interact with one another forming a complex pattern of positive and negative feed backs, making it difficult to simulate, predict, or even quantify ozone exposure. The incidence of ozone induced injury 91 may therefore, be the result of combination of ozone levels and other environmental variables (VanderHeyden et al. 2001). Measurements of closely linked variables such as soil moisture and solar radiation were not measured in this study, and the interaction between uptake and such factors as temperature and wind-speed are species specific and very intricate. Consequently, the occurrence of ozone-like injury was correlated only with ozone concentration for the purpose of this research. The photooxidant damage caused by the ozone molecule indirectly reduces plant growth (Roper et al. 1989; Amundson et al. 1991). The mechanism causing injury is thought to be an ozone-induced increase in cell permeability, leading to a loss of nutrients through leaching by rain (Ashmore et al. 1985) thereby limiting photosynthetic capacity, and reducing overall plant health and productivity. Acute exposure may cause cell death (necrosis) and lead to reductions in photosynthesis and transpiration without any visible injury symptoms. Necrotic tissue does not photosynthesise and if necrosis affects guard cells, stornatal closure will occur (Hil l and Littlefield 1969). Hogsett et al. (1985) found ozone to cause reductions in: stem diameter, plant heights, root dry weight and needle length and number in slash pine seedlings. Internal leaf response to ozone is largely determined by leaf conductance (Reich and Amundson 1985) and hence the number of stomates or reaction sites. The number of (ozone) reaction sites per unit leaf area is also dependent on leaf thickness and hence volume, since the entire leaf cuticle (not just the stomates) may be permeable to the ozone molecule (Kerstiens and Lendzian, 1989). This may mean that conifer needles can take in more ozone than deciduous leaves, due to a higher surface area to volume ratio, making them more at risk to ozone damage. The diffusion of ozone into plant tissue, and hence ozone uptake by vegetation 92 is also dependant on the ozone concentration inside the leaf cells (Krupa and Kickert 1997). Ozone uptake may therefore decrease, as cells become 'ozone saturated' throughout the growing season. Ozone possesses the ability to cause damage to cellular walls and protective articular waxes making an injured plant more susceptible to damage from other chemicals or pathogens (McLaughlin etal. 1998). Since plants take up ozone direcdy through stomates anything that affects stomatal conductance or resistance also effects uptake and hence ozone dose and exposure. Water stress causes plant stomates to close in order for moisture to be conserved. This decreases to flux of ozone into plants in areas or times of low moisture (Dobson et al. 1990; Beyers et al. 1992; Schaub et al. 2003). The flux of ozone through a plant is therefore greater for watered than unwatered plants. However, in some individuals, water stress may lead to a reduction in natural defences and make a plant more susceptible to ozone injury (McLaughlin and Downing 1995; Paakkonen et al. 1998). Stomatal resistance which regulates ozone uptake, increases with radiation intensity and low humidity, the same factors that increase ozone formation by photodissociation of nitrate. Therefore mid-range ozone concentrations may be more damaging to plants than high levels or episodes which would correspond to stomatal closure and thus reduced uptake and injury protection (Griinhage and Jiiger 1994). Ozone itself may enhance stomatal uptake by instigating stomates to open (Dobson et al. 1990), leading to a positive feedback between ozone concentration and rates of ozone uptake. Uptake of ozone by understorey plants has been found to be more significant than uptake rates of water vapour (Lamaud et al. 2002). Similarly, conifers may have ozone fluxes up to three times the gas exchange rate of water Vapour (Rodon et al. 1993). This makes 93 internal leaf sites potentially significant canopy-scale ozone sinks (Lamaud et al. 2002). Since uptake of ozone by plants is controlled by such a myriad of factors, different genotypes of the same species may also take up and respond differendy to ozone exposure (Fuhrer eta/. 1997; Krupa and Kickert 1997). For instance, genotypes at varying elevations may differ in uptake and sensitivity due to different mechanisms developed for survival in the face of various stressors such as low moisture, intense sunlight, or high winds. Apart from the indirect effects of ozone on plant productivity and growth, there are other more direct invisible effects of ozone on vegetation. Ozone may cause alterations in plant performance and competition leading to an alteration in community structure (Bergmann et al. 1999), where only the resistant survive. Pinus strobus pollen exposed to ozone under moist conditions (such as a forest floor) has a reduced germination frequency (Benoit et al. 1983) and this may strongly effect the survival of this ozone sensitive species in high ozone regions. Other Pinus spp. (pries) may suffer from similar germination effects. There may be other influences of ozone on forest communities that are not a function of stornatal uptake - such as the oxidation of soil nitrite. As mentioned in the previous chapter, ozone uptake by vegetation is strongly influenced by high temperatures and light intensities (Coe et al. 1985; Rodon 1993). Padro et al. (1991) found ozone deposition velocities above a deciduous forest canopy to be ten-fold higher in the daytime than at night These are the same conditions and diurnal patterns, that govern photochemical ozone production and the high concentrations observed in the L F V during summertime periods characterised by synoptic-scale high-pressure systems (McKendry 1994; McKendry et al. 1998; Comrie 1990). Rates of uptake are governed by stornatal resistance and therefore follow a diurnal pattern. However, in the afternoon leaf surface 94 warming generally leads to stomatal closure in effort to reduce moisture loss. However, these same hours correspond to the highest rate of ozone uptake in coniferous species (Coe et al. 1995), since ozone uptake is enhanced by high levels of sunlight Solar radiation may be up to three times more efficient at regulating ozone uptake, than stomatal resistance alone (Coe et al. 1995). Musselman and Minnick (2000) suggest the use of 24-hour data as used here when calculating cumulative exposure metrics since vegetation may be particularly susceptible to ozone injury at night Salardino and Carroll (1998) also found metrics calculated using 24-hour data to be well correlated with injury. New evidence shows that gaseous uptake rates may be high even during the night-time (Emberson et al. 2000; Musselman and Minnick 2000) effecting the rates of dark respiration similar to the effects of ozone on photosynthesis (Reich 1983). Therefore, the calculation of cumulative indices from 24-hour, rather than daylight-hour data is appropriate since the ozone can accumulate within / injure a plant cell at any time of day. Cumulative damage indices should include data from any time when uptake occurs even though rates are variable. 6.4 Summary of Impacts on Forest Vegetation Canada's air quality standard for ozone is 82 ppb (Stull 2000) and was formulated to protect human health, rather than plant health, and plants are known to be sensitive to ozone at much lower concentrations. However, even the high 82ppb standard was exceeded on five occasions at site 4 during the episode from Aug. 8-17, 2001. Correlation between plant injury and ozone exposure alone can not be made since neither are independent variables (Kickert and Krupa 1991). It is therefore difficult to set standards to protect vegetation based on solely ambient concentration. Any protection standard 95 / model needs to be based on actual plant uptake of ozone, and include such variables as atmospheric and leaf resistances (Grunhage and Jager1994). For instance, studies have reported that phytotoxic potential may be greater at the mid-range concentrations (50-90ppb) than at higher concentrations (>90ppb). This is because higher concentrations are associated with warm temperatures, high insolation and low atmospheric humidity (Grunhage and Jager 1994). These same weather conditions lead to an increase in stornatal resistance and hence a decrease in stornatal uptake of ozone. Between and within species genetic variability also aids in the difficulty of correlation between ozone concentrations and injury direcdy. It is also arduous to diagnose injury in the field as being ozone inflicted, due to the innumerable potential other stressors also inflicting injury. In addition, many environmental factors contribute to plant sensitivity or resistance to ozone, at either the species or population levels. Chronic injury would increase with elevation since ozone levels are consistently high at these locations and they also represent regions of high peak exposures, such as during the August 2001 episode. Mil ler and Arbaugh (2000) found the incidence of ozone injury to increase with elevation and in this study nearly all shrub species at site 4 illustrated some signs of potential injury. However, at high elevations it is difficult to aistinguish between injury symptoms caused by ozone, and those indicative of other stressors such as drought, wind or frost (Reich et al. 1985). Also it is important not to make conclusions based on visible symptoms alone since invisible damage to growth and biochemical pathways often occurs in the absence of visible injury (Wang and Bormann 1986). Despite all of the limitations in both diagnosing injury, and correlating it to ozone, the injury found at the study sites is indicative of ozone injury. Therefore, the forests on the south-facing slopes of the Lower Fraser Valley may be at risk of oxidant damage. Damage may not 96 only reveal itself as visible foliar symptoms, but also as reductions in growth and productivity, reduced pollen viability, and alterations in forest composition. Species composition may change as a result of ozone resistant species or genotypes, replacing those sensitive to ozone. Genera in the L F V that appear especially ozone sensitive are Acer and Vaccinium spp. which may prove useful regional bioindicators of high ozone. Only one tree was included in the injury monitoring, as it is a coniferous forest and injury to conifers is not only difficult to diagnose, but often not visible as foliar injury (Brace et al. 1999; Innes et al. 2001). In conclusion, although there is potential damage occurring to economically and ecologically important species on the south-facing slopes of the LFV , it is impossible to say without a doubt that the damage is significant or that ozone is responsible. 97 7.0 C O N C L U S I O N S The following includes a summary of findings from the preceding chapters and discusses how spatial and temporal variation in tropospheric ozonemay effect forest susceptibility to oxidative damage in the L FV . Recommendations for future research are also made in order to answer some of the questions regarding ozone's threat to the forests of the L FV . This chapter concludes with a discussion of how these and future results may be applied to forest management and air quality initiatives, and suggest implications of ignoring, tropospheric ozone as a threat to forest ecosystems in the region. 7.1 Summary of Findings Although the city of Vancouver, where most photochemical smog is produced, has relatively clean air, urban pollution is transported eastward to rural areas where ecological damage may be occurring. In the Lower Fraser Valley, summer maximum ozone concentrations may exceed the National Ambient A i r Quality Objective of 82 ppb for extended periods. These elevated ozone levels pose a threat to not only human health, but also to the health of forests in the area. The high maximum values may mean that the forests located at these study sites are at a risk of being damaged by ambient ozone levels. For much of the season, hourly average ozone concentrations may remain quite low. However, the majtimum values at each site were high, thereby indicating the occurrence of acute ozone exposure. For example at all four study sites, ozone exceeded 60ppb for a number of hourly averages throughout the period from August 10* - 17*, 2001, and ozone is known to cause damage to plants at levels as low as 40ppb (Krupa and Kickert 1997). Levels during episodes are therefore deemed high enough 98 to cause injury to vegetation. Diurnal variation in ozone concentration decreases with altitude, while both mean and cumulative ozone levels increase with altitude. This makes sub-alpine forests particularly at risk of oxidant damage as they get a larger pollutant dose (Puxbaum er al. 1991; Sandroni et al. 1994; Zaveri et al. 1995). With in a low forest canopy there was found to be a strong vertical gradient in ozone concentration (9.3 - 7.4 ppb) averaged over two-week monitoring periods. The gradient shows a power-law decrease in ozone concentration towards the surface (forest floor). The two-week average is considered representative of within canopy profiles since concentrations in forest canopies may not show the pronounced diurnal cycle of sites located in clearings continuous (Karlsson et al. 2002). The strong vertical gradient in ozone concentrations is due to a combination of factors including lack of nocturnal down-mixing, dry and wet deposition at the surface, and most importantly uptake by vegetation. Furthermore, the power-law relationship suggests that forest plants may be acting as effective ozone sinks in the region. It was initially suspected that differing concentrations within the canopy would effect the ozone dose received by vegetation of various heights and foliar densities, thereby influencing risk of injury. However, carbon assimilation is greatest at the canopy top where ozone concentrations are the highest (Ollinger et al. 1997), therefore reductions in forest productivity and growth are not affected by the power-law relationship. Conversely, if the decrease in ozone concentration towards the forest floor is due to uptake, then injury may be the greatest at low ozone canopy levels. Although neither the A O T 4 0 or SUM06 metric were exceeded in either 2001 or 2002, symptoms indicative of ozone injury were found on native forest shrubs at the study sites in 2002. The frequency and severity of these symptoms increased with elevation, and hence 99 cumulative ozone exposure. A t this time it is impossible to diagnose without a doubt, the injury as being ozone induced, however levels are deemed high enough to cause visible injury, especially at high elevations. Furthermore, the plants displayed classic ozone injury symptoms that worsened with prolonged exposure and included premature leaf senescence and abscission. This research was thorough in the sense that it examined and discussed ozone and forest interaction at many scales - from synoptic circulation and seasonal climate to cellular response and stomatal function. However, this study was in no way complete, and unfortunately answered very few questions regarding how current levels of tropospheric ozone are effecting forests in the LFV . Even though the first three objectives posed in chapter 1.0 were met, and the fourth objective (vegetative injury) explored, it is still not possible to make any definitive statements regarding the threat of injury that ozone poses to forests of the LFV . Henceforth future research is required to understand the complex multi-scale interactions between ozone and forests in this region. 7.2 Future Research Continuous long-term monitoring of tropospheric ozone in the LFV , is required in order to fully assess the dose of ozone to forest vegetation. It also becomes necessary in determining whether or not the levels are high enough within the valley to pose a significant threat to forests over the long-term. Monitoring should include hourly average ozone and meteorological variables from sites of various elevations, as used in this study. However it would be useful to also monitor these variables at elevations of 800m and 1000m. This would answer questions of how ozone varies with elevation under different boundary layer conditions, 100 and potentially explain the seasonal differences in elevational patterns found between 2001 and 2002. Unti l now it was unknown how tropospheric ozone concentrations vary within a forest canopy. It is still unknown what factors are effecting the relationship, or how these concentration gradients effect ozone dose to vegetation and hence injury. Wi th in canopy monitoring in the future should include hourly averages at various heights so that diurnal cycles in ozone concentration (if significant) may be applied to diurnal uptake patterns. Hourly average ozone concentrations may be obtained by the use of a continuous monitoring system or by repeat passive sampling with the inclusion of a pyranomter to measure short-wave radiation. Cumulative ozone concentrations may then by converted to hourly averages using a model developed by Krupa et al. (2003). It is also necessary to compare forest-canopy results with an open field in the same area in order to distinguish between the relative influence of deposition versus uptake and stable versus unstable or neutral conditions. Ozone flux towers in both open field and forest canopy environments would provide extensive insight into deposition velocities and uptake rates and the relative dose and threat of ozone to various species at differing canopy heights. Flux towers combined with techniques of eddy correlation would also determine how different stability regimes affect vertical ozone profiles and ascertain the extent of vertical mixing. So far, dose-response models only work for the data utilised to create die model (Krupa and Kickert 1997). Multi-scale models need to be developed in the future before predictions of forest damage can be made. These models need to include such things as stornatal fluxes rather than just ozone exposures, since conductance may prove a more reliable measure of plant exposure than concentrations alone (Emberson et al. 2000; Karlsson et al. 101 2002). Ollinger et al. (1997) coupled dose-response data, with forest-ecosystem model (PnET) to predict ozone effects on forest productivity using leaf-, canopy- and stand-level processes. The coupling of multi-scale models, due to the complex nature of forest-ozone interactions, should be included in future forest injury research. Models should also account for the random distribution of resistant or susceptible genotypes within a species' population, governed by genetic diversity. It may be useful to calculate the survival probably in terms of ozone exposure for those species that show premature senescence. Also a more detailed assessment of seasonal development of injury on particular leaves over the growing season, rather than an entire plant would provide more detailed injury information (Ghosh et al. 1998), One way of establishing a relationship between ozone impacts and forest health would be to correlate exposure with a parameter such as stem density or crown growth in a long-term monitoring project (Karlsson et al. 2002). The development of critical levels that are species' and or plant specific to areas of the L F V would aid in the prediction / long-term modelling of forest injury. These critical levels should include factors such as changes in species; or genotypic frequency within populations (Bergmann et ai. 1999). Ozone resistance may be of a high enough frequency in natural populations that but-crossing may provide sufficient tolerance in some species (Miller et al. ' 1983), however some species may be particularly sensitive to injury and thereby selected against in ozone rich ecosystems. Open-top chambers and fumigation studies can be done on native in situ forest vegetation to determine species' or genotype specific critical levels. This would be especially valuable for Vaccinium and Acer species to determine whether or not the foliar injury found in the L F V was indeed ozone induced. In areas with particularly high 102 concentrations of ozone, such as at high elevations, individuals and species should be examined for apparent resistance and sensitivity. Through molecular biological techniques genetic diversity and even the frequency of resistance may be determined. Only through continued monitoring and an interdisciplinary multi-scale modelling approach can forest growth reductions and ecosystem alterations by ozone be predicted or determined. The threat exists, and thus the tools to understand the implications of ozone injury need to be developed before any policy initiatives will take place. 7.3 Applications and Implications If the symptoms recorded and examined in this study, are indeed effects of ambient ozone concentration, Vaccinium and Acer spp. may prove good bioindicators of oxidant damage. These genera suffered more severe symptoms, occurring more frequendy and symptoms were identical to foliar injury by ozone recorded in previous studies. If these two genera or species within them, have distinctive sensitivities they may be used in conjunction to relate exposure to peak values or pro-longed exposure (Bergmann et al. 1999). Biomonitoring programs using regionally specific indicator species can be implemented as an economically efficient way of monitoring air quality and knowing when critical levels are being exceeded. These projects can include participation from interested members of the public, and may be used as early warnings of declining air quality in areas affected by high levels of ozone. Ozone has very direct effects on ecosystem productivity and function (Sanz and Mil lan 2000). Changes in the hydrologic cycle may prove early warning indicators of whole ecosystem response to ozone damage (Kickert and Krupa 1990). Changes in the hydrologic cycle 103 instigated by ozone may adversely affect water quality and availability thereby posing an escalated threat to ecosystems and society. In a model developed by Kickert and Krupa (1990) the L F V is projected to have increasing ozone concentrations that will be high enough to cause severe ecosystem damage by the year 2025. W o o d growth reductions by ambient ozone of 3 - 22% may occur in forests (Olhnger et ai. 1997) where ambient levels are high enough to cause injury. Future increases in ozone will effect British Columbia's economy drastically, as softwood lumber is the province's most lucrative industry, and may result in economic losses of billions of dollars in the next quarter century. By ignoring ozone as a threat to forest in the LFV , the region becomes vulnerable to severe economic and environmental damage. This project is the first to explore the effects of tropospheric ozone on the forests of Western Canada. Partnerships with both European and American governmental and academic communities may allow Canada to implement policy governing air quality and pollutant emissions before forest injury comparable to that of other nations' occurs. Atmospheric pollutants, including ozone, are trans-boundary problems and therefore require trans-boundary solutions. Nevertheless, each region is very specific in terms of its response to ozone and research must be conducted on a provincial or even municipal level. Additionally, emission standards and alternative forms of energy that produce fewer ozone precursors need to become widespread and implemented more rigorously if we wish to ensure the health, vitality, diversity and economic benefits of our forests for future generations. 104 A P P E N D I X A - L A B O R A T O R Y P R O T O C O L (Modified from Harvard School of Public Health, 2001) A . l Introduction The ozone passive sampler badge.is a passive monitoring device that was developed by Ogawa & Co. for collecting nitrogen oxides (NOx). The filter used for ozone monitoring is coated with a nitrite-based solution with which ozone reacts to oxidize the nitrite (NO2) to nitrate (NOs). Following exposure to ambient conditions, the filter's coating is extracted with ultra-pure (Milli-Q) water. The extract is analyzed using ion chromatography (IC) to determine the nitrate ion concentration, which is then used to calculate the cumulative amount of ozone collected of the exposure period. A batch of samplers consists of both field blanks and field samples, which share the same coating date, the same preparation date, the same handling, and preferably, the same IC analysis date. Since unexposed filters are stored refrigerated before being loaded into samplers, all filters comprising a batch must be removed from refrigeration at the same time. Samplers of a common batch are assembled, exposed, and disassembled over a designated period as defined by the study. A t least 10% of the samplers from each batch should be field blanks. The blanks are transported with samplers to the field site. Each blank is handled similarly to the samples, and is removed from the storage bottle and the resealable bag, exposed to ambient conditions, and then immediately placed back in the bag and in the bottle. Exposed samplers should not be refrigerated during shipping (or at anytime prior to 105 preparation for IC analysis) because humidity within the resealable bag may condense and wet the filter. A.2 Assembly of Passive Samplers for use in the field A.2.1 Equipment and Supplies: 1 sampler body (with 2 spacer disks & 2 rings) 2 diffusion end-caps (1 per end) 4 stainless steel screens (2 screens per end) 1 pin-clip holder 1 storage bottle (amber polystyrene) 1 resealable plastic bag tape to seal storage bottles ID (identification) labels (2 per sample) pre-coated filters for ozone (2 per sample) 1 forceps, blunt for filter handling 1 forceps, sharp with curved tip ethanol lint-free paper wipes (Kimwipes or equivalent), large & small 1 plastic squeeze bottle with MilU-Q water A glove box was not used in this procedure but the surroundings were kept clean and lint free while the work was done quickly to limit fdter exposure to air. 106 A.2.2 Samler Assembly 1) Make sure all parts are clean and dry (see A . 7) 2) Lay out some clean kimwipes on the lab bench 3) Place no more than 25 sampler barrels (one end up) on the kimwipes - may use hands as long as only barrel sides are touched 4) Using clean forceps (rinsed in distilled and deionized water and dried with a kimwipe) place a flat white spacer disk into each barrel (one side only) 5) Using clean forceps place a white ring onto each spacer disk (one side only) 6) Using clean forceps place a stainless steel screen onto each white ring (one side only) 7) Carefully place an unexposed filter onto each stainless steel screen - it is important to have removed filters from refrigerator prior to this and to ensure that they have wanned up to room temperature. Minimise the time that filters are exposed to air. This step and subsequent steps should be done as quickly as possible to minimise filter exposure. 8) Using clean forceps place another stainless steel screen on top of each filter (one side only) 9) Using thumb and forefinger grasp the edge of the Teflon end plug (the piece with the holes in it) and snap into place on one end of sampler. 10) Invert all completed samplers 11) Repeat steps 3-8 on second side of sampler 12) Place loaded sampler barrels into sampler holder clip. 13) Place sampler in clip into plastic bags, seal bags and place into amber vials. Place cap on vials and attach duplicate labels. 107 Note: Pre-coated filters are supplied by Ogawa & Co., USA , Inc. in vials labeled with the date of coating. A laboratory record is kept of the date of filter coating and the date of assembly for each batch of samples and blanks. The pre-coated filters must be stored in their original containers in a cool, dark place, preferably at 5°C. A.3 Passive Sampler field exposure 1) For sampling place one label on sampler body and a second label on the amber botde. 2) Cl ip sampler into rain cover and mount in a desired monitoring area. 3) Make a note of the time and date of the sampling initiation. 4) T o end sampling - remove the badge from the sampling location and place it in the resealable bag, then in the storage bottle (this was done after two weeks of exposure). 5) Be sure to leave the ID label on the sampler, and tape the cap securely on die bottle. 6) Make a note of the sampling end date and time in the field data log. A.4 Disassembly of Passive Samplers after field exposure A.4.1 Equipment clean, dry extract vials with caps beakers for used sampler components small and large Kimwipes Mi l l i -Q water calibrated automatic dispensing pipette (5 ml) forceps, blunt for filter handling forceps, unserrated, sharp with curved tip 108 A.4.2 Sampler Disassembly 1) Remove the sampler from the protective bottle and resealable bag. 2) Remove the sampler body from the pin-clip. 3) Select a clean extract vial. 4) Remove the label from the pin-clip and transfer it to the extract vial. 5) Remove die cap from the extract vial. 6) Dispense 5ml of Mi l l i -Q water into the extract vial using the automatic dispensing pipette 7) Ho ld the sampler body over a clean Kimwipe on a tray. Ho ld the sampler body using fingers, with one end pointing up. Remove the top end-cap of the sampler body (use clean flat forceps as a wedge, if necessary) and place the end-cap in a beaker designated for used end-caps. 8) Still holding the sampler body over the tray, tilt it and using forceps remove the first screen and then the filter, being careful not to damage the screens. 9) Place the filter into the extract vial, if the filter does not insert easily, fold the filter using the two forceps together. 10) Remove the second screen and all other sampler parts, placing them in designated beakers. 11) Remove the filter from the other end of the body using the same techniques as above, and place it in the same vial as the first filter. Cap the extract vial securely. 12) After finishing transfer of both filters to the extract vial, clean the forceps, with moist Kimwipes, making sure that they are all wiped completely dry after cleaning. 13) Repeat the process for each sampling badge. 109 14) Store the extract vials for each sampler in a cool place (but not refrigerated), in the dark, until the time of filter extraction (see below). A.5 Preparation of Passive Ozone Samples for Ion Chromatography (IC) A.5.1 Equipment Forceps, not serrated, sharp with curved tip IC vials (0.5 ml for Dionex auto-sampler) Caps for I C vials , Syringes, 3 ml disposable Mi l lex-LCR13 syringe filters Ultrasonic bath (sonicator) Support rack for extract vials A.5.2 Filter Preparation 1) Check that the filters in each extract vial are completely immersed in the aqueous solution. 2) Place the extract vials in the ultrasonic bath. Adjust the level of water in the bath to be high enough to immerse the bottom of each vial (about 2cm), but not so high that the vials float 3) Sonicate the extract vials for 5 minutes then rotate them 90 degrees clockwise; repeat the rotation twice for a total sonication time of 15 minutes. 4) Separate all the extract vials (blanks and samples) into groups of 10 or so vials each. One syringe/Millex filter combination will be used for each group of 10. 110 5) Withdraw 3ml of Mi l l i -Q water into a sterile syringe 6) Attach a Mil lex filter to syringe end and expel water while firmly holding filter on syringe 7) Separate the syringe from the Mil lex filter 8) Withdraw 1 ml of sample extract from the extraction vial into the syringe. 9) Attach the syringe to the Mil lex filter. 10) Holding the filter firmly onto the syringe, expel exacdy 5 drops into a waste container. 11) Inject 0.5 ml of the sample into an i.e. vial. 12) Discard the remaining sample to waste. 13) T o clean syringe and Mil lex filter between samples by: a. separate the syringe from the filter. b. withdraw 3 ml of Mi l l i -Q water into the syringe. c. reattach syringe to filter, firmly hold filter onto syringe, and expel water into waste. d. repeat steps a-c twice, then return to 1). 14) After using a syringe/Millex filter combination for ten samples, discard both. A.6 Ion Chromatography Analysis A.6.1 Ion Chromatography Instruments and Eluants These are the recommended conditions and components for IC analysis using the Dionex IC system. The exact conditions used may be modified as necessary. I l l A.6.2 IC Instrument Dionex Model 2000i with conductivity detector: anion eluant flow 1.7 mlVnruTiute; regenerant pressure 10 psi; nitrogen 99.9% pure @ 100 psi; eluant pressure 5 psi; detector range 10 uS. A.6.3 Columns Separator column #AS4A (Cat 1137041); guard column I IAG4A (Cat # 37042); anion micromembrane suppressor Model #AMMS - 1 . A.6.4 Eluant for ozone extract (prepared with reagent grade chemicals) 1.08 m M NasCOs, 1.02 m M N a H C O s 12m l0 .36M NasCOs 12ml 0.34M NaHCOa 3976 ml ml Milli-Q-water total volume = 4 L Eluant stock solutions: 0.34 M NaHCOs 28.6+/-0.1 gNaHCOs Mi l l i -Q water total volume = 1 L 0.36 M N ^ C O , 38.16gNa .COs Mi lh-Q water total volume = 1 L 112 Nitrite, nitrate, and sulfate ions are measured using the ion chromatography configuration oudined above. In ozone filter analysis, nitrite is a very large peak since it is a main component of the coating solution. T o measure the nitrate peak quantitatively, it is necessary to adequately separate the nitrite and nitrate peaks. The concentration of the anion eluant may be varied (diluted) to get optimum results. After repeated use, the columns ability to separate the peaks deteriorates, and more dilute eluant is required. A.6.5IC Standards Standards can be prepared for analysis of different anions for different types of samplers, which contain nitrite, nitrate, and sulfate ions. The solution concentrations, in ppm, are given below. For analysis of ozone samples only, these standards can be prepared with only nitrate ion. Standard / Sulfate / Nitrate / Nitrite A N 1 / 0.20 ppm / 0.16 ppm / 0.16 ppm A N 2 / 0.40 ppm / 0.32 ppm / 0.32 ppm A N 3 / 0.80 ppm / 0.64 ppm / 0.64 ppm A N 4 / 2.0 ppm / 1.6 ppm / 1.6 ppm A N 5 / 4.0 ppm / 3.2 ppm / 3.2 ppm A N 6 / 10.0 ppm / 8.0 ppm / 8.0 ppm A.6.6Data Analysis Ion chromatography yields results in the form of NOx ratios. These ratios were converted to cumulative ozone exposure in ppbh over two week sampling periods. The conversion 113 was done using a linear equation formulated by Brauer and Brook (1995) since the regression of passive versus continuous data in this project had a weak correlation due to non-coincident gaps in data. A.7 Routine Cleaning of Sampler Components 1) Disassemble the samplers and set aside the cylinder bodies (with spacer disks and rings still inside) to be cleaned separately. 2) Rinse the end-caps with Mi l l i -Q water, then set the parts on Kimwipes to dry. It may be necessary to tap the water out of the holes in the end-caps in order for them to dry completely. 3) W ipe the cyhndrical bodies clean with Kimwipes moistened with Mi l l i -Q water. Then use a dry Kimwipe to wipe off excess water. 4) Place the stainless screens in a beaker and rinse them several times with Mi l l i -Q water. Fil l the beaker again with Mi l l i -Q and place it in a sonication bath. Sonicate the screens for 5 minutes, then rotate the beaker a quarter turn; this procedure must be continued for total of 15 minutes. 5) A l l parts of the ozone passive sampler must be clean and completely dry before assembling the sampler. After cleaning, lay the parts on Kimwipes to dry. 6) Whi le drying, cover with one large Kimwipe to prevent dust/dirt from settling on the clean parts. 7) Inspect the pin-clips for obvious dust, dirt, etc. If necessary, rinse with Mi l l i -Q water and lay on Kimwipes to dry. 114 A P P E N D I X B - H O U R L Y O Z O N E C O N C E N T R A T I O N S B. l Summer of 2001 T ime Series The following pages contain time series data of ozone concentrations measured at each of the four sites during the 2001 study season. Ozone concentrations are presented as hourly averages in parts per billion by volume (ppb). Each figure represents a two-week measurement represented by Julian day. The Julian days for which there is no / little data were left in the Appendix to maintain a standard time series scale. 115 Days 176-189 -Site 1 -Site 2 •Site 3 •Site 41 < ^ <!> # # v # ^ ^ ^ v<£ v # ^ V V V V V V V V V V V N° V V v v v» V Julian Day 90 Days 190-203 •Site 1 -Site 2 •Site 3 -Site 4 # # # ^  ^  ^ $ ^  $ $ ^  ^  ^  ^  ^  ^  ^ Julian Day 116 Days 204-217 90 -j c 80 ntratio ntratio 70 -u Cone 60 # # <& # 4 ^ 4" ^  4" & & «>fa & Julian Day Days 218-231 cj>* cj? cJJ? # cf> c?> # # c$> # # ^ # ct Julian Day 117 Days 232-245 90 i | 80 -I 70 Julian Day Site 1 Site 2 Site 3 Site 4 118 Days 260-273 Julian Day 119 A P P E N D I X B - H O U R L Y O Z O N E C O N C E N T R A T I O N S B.2 Summer of 2002 Time Series The foUowing pages contain time series data, of ozone concentrations measured at each of the three sites during the 2002 study season. Ozone concentrations are presented as hourly averages in parts per billion by volume (ppb). Each figure represents a two-week measurement represented by Julian day. Recall that the data from sire 2 was discarded due to instrument error and that there was data loss at each of the sites for often extended periods (especially in AugusO. The Julian days for which there is no / httle data were left in the Appendix to maintain a standard time series scale. 120 Days 186-199 u N CU 0 3*60 u a | | 50 < g40 la 3 0 10 0 <o i>» oo <y> 00 00 00 X - H CN CO "5 0> 0> O} 0> Oi t>> CO Ol 0> Oi O; Julian Day 121 100 90 80 Days 200-213 •Site 1 •Site 3 •Site 4 O O O O O O O O O O O - H ^ — ^ r - i CN CM CM CN CN OI CM CN CM CM CN CN CN CM Julian Day 100 90 80 Days 214-227 -Site 1 •Site 3 -Site 4 • ^ " ^ i o t o r > . o o c ? i O - ^ C M c o - * l o t o c ^ ^ ^ r t ^ ^ ^ ^ C N C M C M C M C M C M C M C M CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM Julian Day 122 100 90 80 Days 228-241 Site 1 •Site 3 •Site 4 C O O O O i O ' - H C N I C O ^ ^ c O C ^ O O C n CN CN CM CN CN CN C M C N C ^ C N C N C N C N C N Julian Day 100 90 80 3 -S N g, 0-^60 u a ft* 2 ^ § 4 0 53 20 10 Days 242-255 •Site 1 •Site 3 •Site 4 - i 1 1 1 r CN CN CO *f3 <0 TJ< TJ* CN CN CN CN CM CN CN CO O i CN CO ^ «5 «! >fl CN CN CN CN CN CN CN CN Julian Day 123 u c I 0 Sb B u < 3 0 100 90 80 s 70 a 3 6 0 150 CZ B i 4 0 Jso 20 10 0 Days 256-269 •Site 1 •Site 3 •Site 4 i 1 1 1 r < 0 < O t > « C O O > © » - H C N e o , , < ! j , « } < © t > « 0 0 , 0 > CM OI O) CM OI CM Ol CM OI OI OI CM O} CM Julian Day Days 270-281 u c SP 0 > -O < c >-, u s o O 70 60 50 40 30 20 10 0 © © r H t N tN . t > CM CM CM CM CO "tf »0 <0 00 r>* c-x r>> o t-^  CM CM oq CM CM CM CM C?> © co co CM CM CM Julian Day 124 APPENDIX C - V E G E T A T I O N IN JURY C . l Photos and Horsfall-Barratt Scores The pages of the following section contain photos (taken digitally) of each of the individuals plants monitored throughout the 2002 field season. The top of each page provides the site where the plant was located and its' species name. Below each photo is the date of image capture and the corresponding Horsfall-Barratt score (HB). See section 3.4 and Appendix C.2 for HB-score calculation. Note that some individuals show additional injury that does no appear to be caused by ozone. This unrelated injury is not discussed in the text or included in the HB-score. 125 Site 1 - Rubus parviflorus Site 2 - Acer circinatum September 3, HB-5000 September 17, HB-7500 133 Site 3 - Cornus canadensis Site 3 - Vaccinium parvifolium b) September 3, HB-1500 September 17, HB-8800 136 Site 3 - Acer circinatum September 3, HB-4500 September 17, HB-5250 137 Site 3 - Vaccinium membranaceum September 3 - dead September 17- dead 139 Site 4 - Vacinnium membranaceum a) Site 4 - Vacinnium membranaceum b) Site 4 - Vacinnium membranaceum c) Site 4 - Vacinnium ovalifolium a) August 27, HB-570 143 Site 4 - Vacinnium ovalifolium b) Site 4 - Vacinnium ovalifolium c) A P P E N D I X C - V E G E T A T I O N INJURY C.2 Injury Symptoms and Severity The following section contains a table describing the severity of potential ozone injury symptoms found on individuals in the summers of 2002. Individuals are listed by site location - Arboretum, site 1 (200m), site 2 (400m), site 3 (600m) and site 4 (1200m), and by species' name. The date of qualitative data collection is given, and a corresponding visual estimate of percent leaf area (LA) injured and percent of total plant foliage (TP) affected, is given. These two estimates are than multiplied to give a Horsfall-Barratt (HB) injury score (see section 3.4). 146 Arboretum Spirea douglasii Vaccinium parvifohum Date L A (96) T P (96) H B L A (96) T P (96) H B 8-Aug 5 25 125 20 50 100 20-Aug 40 50 2000 10 75 750 3-Sep 40 75 3000 20 75 1500 17-Sep 60 75 4500 40 75 3000 Site 1 Rubus parviflorus Vaccinium parvifohum Date L A (96) T P (96) H B L A (96). T P (96) H B 8-Aug 5 6 30 20 25 500 20-Aug 5 6 30 20 25 500 3-Sep 5 6 30 determined not be be ozone induced 17-Sep 10 6 60 5 12 60 new ozone-like symptoms Site 2 Psuedotsuga menziesti Gaultheria shallon Date L A (96) T P (96) H B L A (96) T P (96) H B 8-Aug 95 50 4750 15 75 1125 20-Aug 100 50 5000 20 50 1000 3-Sep 100 75 7500 50 50 2500 17-Sep 100 75 7500 20 50 1000 Vaccmium parvifohum Acer circinatum L A (96) T P (96) H B L A (96) T P (96) H B 8-Aug 30 88 2640 100 50 5000 20-Aug 20 75 1500 100 .50 5000 3-Sep 20 88 1760 100 50 5000 17-Sep Dead 100 75 7500 Site 3 Gaultheria shallon Vaccinium membraceum Date L A (96) T P (96) H B L A (96) T P (96) H B 8-Aug 20 50 1000 . 60 75 4500 20-Aug 30 50 1500 60 88 5280 3-Sep 30 75 2250 Dead 10000 17-Sep 40 50 2000 Vaccinium parvifohum 1 Vaccinium parvifohum 2 L A (%) T P (96) H B L A (96) T P (96) H B 8-Aug 10 75 750 40 75 3000 20-Aug 30 50 1500 40 75 3000 3-Sep 100 75 7500 30 50 1500 17-Sep 60 75 4500 100 88 8800 Acerglabrum Comus canadensis L A (96) T P (96) H B L A (96) T P (96) H B 8-Aug 50 50 2500 90 ; 50 4500 20-Aug 90 25 2250 90 50 4500 3-Sep 60 75 4500 Green - new leaves 0 17-Sep 70 75 5250 147 Site 4 Vaccinium membraceum a) Vaccinium membraceum b) Date L A (96) T P (96) H B L A (96) T P (96) H B 8-Aug 100 25 2500 . 100 25 2500 13-Aug 100 50 . 5000 100 50 5000 27-Aug 100 25 2500 60 12 720 10-Sep 100 50. 5000 100 50 5000 24-Sep 100 75 7500 100 50 5000 Vaccinium membraceum c) Vaccinium ovalifolium a) L A (96) TP(%) H B L A (96) T P (96) H B 8-Aug 100 25 2500 30 12 360 13-Aug 100 50 5000 30 12 360' 27-Aug 100 6 600 95 6 570 10-Sep 100 75 7500 95 12 1140 24-Sep 100 50 5000 100 . 12 1200 Vaccinium ovalifolium b) Vaccinium ovalifolium c) L A (%) T P (96) H B L A (96) T P (96) H B 8-Aug 30 12 360 30 12 360 13-Aug 30 12 360 30 • 12 360 27-Aug 95 6 370 95 12 1140 10-Sep 90 12 1080 90 25 2250 24-Sep 100 12 1200 100 25 2500 senesence? 148 W O R K S C I T E D Amundson, R.G., R.G. Alscher, S. Fellows, G . Rubin, J . Fincher, P. Van Leuken, and L .H. Weinstein, 1991. Seasonal changes in the pigments, carbohydrates and growth of red spruce as affected by exposure to ozone for two growing seasons. New Phytologist 118: 127-137. Aneja, V . P., Z. L i , and M. Das, 1994. Ozone case studies at high elevation in the eastern United States. Chemosphere 29(8): 1711-1733. Ashmore, M.R., J.N.B. Bell, and A J . Rutter, 1985. Forest Damage in West Germany and the role of ozone. Ambio 14: 81-87. Ball, G.R., J . Benton, D. Palmer-Brown, J . Fuhrer, L. Skarby, B.S. Gimeno, and G. Mills, 1998. Identifying factors which modify the effects of ambient ozone on white clover (Trifolium repens) in Europe. Environmental Pollution, 103: 7-16. Banta, R.M., P.B. Shepson, J .W. Bottenheim, K.G. Anlauf, H.A. Wiebe, A . Gallant, T . Biesenthal, L.D. Oliver, C. Zhu, I.G. McKendry, and D.G. Steyn, 1997. Nocturnal cleansing flows in a tributary valley. Atmospheric Environment 31(14): 2147-2162. Barbo, D.N., A . H . Chappelka, G.L. Somers, M.S. Miller-Goodman, and K. Stolte, 1998. Diversity of an early successional plant community as influenced by ozone. New Phytologist, 138: 653-662. Barbo, D.N., A . H . Chappelka, G.L. Somers, M.S. Miller-Goodman, and K. Stolte, 2002. Ozone impacts on loblolly pine (Pinus taeda L.) grown in a competitive environment. Environmental Pollution, 116:27-36. . Bennett, J.P. and V.C. Runeckles, 1977. Effects of low levels of ozone on plant competition. Journal of Applied Ecology, 14: 877-880. Benoit, L.F., J .M . Skelly, L.D. Moore, and L.S. Dochinger, 1982. Radial growth reductions of Pinus strobus L. correlated with foliar ozone sensitivity as an indicator of ozone-induced losses in eastern forests. Canadian Journal of Forest Research 12: 673-678. Benoit, L.F., J .M . Skelly, L.D. Moore, and L.S. Dochinger, 1983. The influence of ozone on Pinus strobus L. germination. Canadian Journal of Forest Research 13: 184-187. Bergmann, E., J . Bender and H.-J. Weigel, 1999. Ozone threshold doses and exposure-response relationships for the development of ozone injury symptoms in wild plant species. New Phytologist 144: 423-435. 149 Beyers, J.L., G . H Reichers, and P J . Temple, 1992. Effects of long-term ozone exposure and drought on the photosynthetic capacity of ponderosa pine (Pinus ponderosa Laws.). New Phytologist 122: 81-90. Blumentahl, D.L., W . H . White, and T .B . Smith , 1978. Anatomy of a Los Angeles smog episode: pollutant transport in the daytime sea breeze regime. Atmospheric Environment 12: 893-907. Bdhm, M., B. McCune, and T . Vendetta, 1991. Diurnal curves of tropospheric ozone in the western United States. Atmospheric Environment 25A(8): 1577-1590. Brace, S. 1996. The Spatial Distribution of Ozone in the Mount Rainier National Park Region. University of Washington, MSc Thesis. Brace, S., and D.L. Peterson, 1998. Spatial patterns of tropospheric ozone in the Mount Ranier region of the Cascade mountains, U.S.A. Atmospheric Environment 32(21): 3629-3637. Brace, S., D.L. Peterson, and D. Bowers, 1999. A guide to ozone injury in vascular plants of the Pacific Northwest. U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station, General Technical Report PNW-GTR-446. Brauer. M . and J.R. Brook, 1995. Personal and Fixed-Site Ozone Measurements with a Passive Sampler. Journal of Air and Waste Management Association 45: 529-537. Broder, B., H .U. Dutsch, and W . Graber, 1981. Ozone fluxes in the nocturnal planetary boundary layer over hilly terrain. Atmospheric Environment 15(7): 1195-1199. Browell, E.V., M A . Fenn, C.F. Buder, W . B . Grant, R.C. Harriss, and M.C. Shipham, 1994. Ozone and aerosol distributions in the summertime troposphere over Canada. Journal of Geophysical Research 99 (D l ) : l 139-1755. Bul l , K.R. 1991. The critical load levels approach to gaseous pollutant emission control. Environmental Pollution 69: 105-123. Bytnerowicz, A., B. Godzik, W . Fraczek, K. Grodzinska, M . Krywult, O . Badea, P. Barancok, O . B lum M . Cerny, S. Godzik, B. Mankovska, W . Manning, R. Moravcik, R. Musselman, J . Oszlanyi, D. Postelnicu, J . Szdzuj, M . Varsavova, and M . Zota. Distribution of Ozone and other air pollutants in forests of the Carpathian Mountains in central Europe. Environmental Pollution 116: 3-25. Bytnerowicz, A , D .M. Olszyk, S. Huttunen and B. Takemoto, 1989. Effects of photochemical smog on growth, injury and gas exchange in pine seedlings. Canadian Journal of Botany 67: 2175-2181. 150 Cadle, R.D. 1971. The Chemistry of Smog. In Man's Impact on the Climate. W . H . Matthews, W . Kellog and G.D. Robinson (eds.). M I T Institute of Technology, Cambridge, Massachusetts USA . Carlson, C. 1997. Do concentrations of tropospheric ozone in eastern Mpumulanga pose a risk to commercial forests? South African Journal of Science 93(1): 7-10. Coe, H., M .W. Gallagher, T . W . Choularton and C. Dore, 1995. Canopy scale measurements of stornatal and cuticular Os uptake by Sitka Spruce. Atmospheric Environment 29(12): 1413-1423. Comrie, A .C . 1990. The climatology of surface ozone in rural areas: a conceptual model. Progress in Physical Geography 14: 293-316. Cooper, S.M., and D.L. Peterson, 2000. Spatial distribution of tropospheric ozone in western Washington, USA . Environmental Pollution 107: 339-347. Corsmeier, U., N. Kalthoff, O . Kolle, M . Kotzian, and F. Fielder, 1997. Ozone concentration jump in the stable nocturnal boundary layer during a LLJ-event Atmospheric Environment 31: 1977-1989. Dobson, M.C., G. Taylor, and P.H. Freer-Smith, 1990. The control of ozone uptake by Picea abies (L.) Karst. and P. sitchensis (Bong.) Carr. during drought and interacting effects on shoot water relations. New Phytologist 116: 465-474. Drewitt, G . 1996. Measurement of biogenic hydrocarbon emissions from vegetation in the Lower Fraser Valley, British Columbia. MSc Thesis, the University of British Columbia, 121pp. Edmonds, R.L. and F.A. Basabe, 1989. Ozone concentrations above a Douglas-fir forest canopy in western Washington, U.S.A. Atmospheric Environment 23(3): 625-629. Emberson, L.D., G. Wieser, and M.R. Ashmore, 2000. Modell ing of stornatal conductance and ozone flux of Norway Spruce: comparison with field data. Environmental Pollution 109:393-402. Environment Canada, 2003. Climate Trends and Variation Bulletin. Summer 2001 and 2002. http://www.msc-smc.ec.gc.ca/ccm htfp://www.msc-smc.ec.gc.ca/ccm Fast, J.D., and S. Zhong, 1998. Meteorological factors associated with inhomogeneous ozone concentrations within the Mexico City basin. Journal of Geophysical Research 103 . (D15):18 927-18 946. Forest Inventory and Analysis (FIA) 2003. U S D A Forest Service and State Partners Biomonitoring information website www.fiaozone.net. 151 Fishman, J. , C E . Watson, J.C. Larsen, and j . A . Logan, 1990. Distribution of Tropospheric Ozone Determined from Satellite Data. Journal of Geophysical Research 95 (D4): 3599-3617. Fredrickson, T.S., T .E . Kolb, J .M . Skelly, K.C. Steiner, B J.Joyce and J.E. Savage, 1996. Light environment alters ozone uptake per net photosynthetic rate in black cherry trees. Tree Physiology 16(5): 485-490. Fuhrer, J . and Acherman, B. (Editors) 1994. Critical levels for ozone: a U N - E C E Workshop Report. Swiss Federal Research Station for Agricultural Chemistry and Environmental Hygiene, Liebefeld-Bern, Switzerland. Fuhrer, J., L. Skarby, and M.R. Ashmore, 1997. Critical levels for ozone effects on vegetation in Europe. Environmental Pollution 97(1-2): 91-106. G V R D 2003. B C Ministry of Water, Land and A i r Protection web-site. Lower Mainland, Archived Data. www.elp.gov.bc.ca:8000/pls/aqiis/aqi.bulletin Gill iam, F.S. and Turri l l , N.L. 1995. Temporal patterns of ozone pollution in West Virginia: implications for high elevation hardwood forests. Journal of Air and Waste Management Association 45:621-626. Ghosh, S., J .M . Skelly, J.L. Innes, and J.L. Skelly, 1998. Temporal development of visual ozone injury on the foliage of Prunus serotina - a statistical evaluation. Environmental Pollution: 102: 287-300. Glisten, H., G . Heinrich, and D. Sprung, 1998. Nocturnal depletion of ozone in the upper Rhine Valley. Atmospheric Environment 32: 1195-1202. Griinhage, L. and H J . Jager, 1994. Influence of the atmospheric conductivity on the ozone exposure of plants under ambient conditions: considerations for establishing ozone standards to protect vegetation. Environmental Pollution 85: 125-129. Griinhage, L. and H J . Jager, 1996. Critical Levels for ozone, ozone exposure potentials of the atmosphere or critical absorbed doses for ozone: a general discussion. In Proceedings of the Workshop on Critical Levels of Ozone for Europe, Apr i l .1996, Kuopio, Finland. University of Kuopio, Kuopio, Finland. Harvard School of Public Health, 2001. Protocol for ozone measurement using the ozone passive sampler badge, revision 3: February 2001. Source: htrp://www.imlmc.conVairscience/pass.. Heck, W .W. , J.A. Dunning, and I J . Hindaw, 1966. Ozone: Nonlinear Relation of Dose and Injury in Plants. Science 151: 577-578. 152 Hi l l , C. and N. Litdefield, 1969. Ozone. Effect on Apparent Photosynthesis, Rate of Transpiration, and Stornatal Closure in Plants. Environmental Science and Technology 3(2): 52-55. Hogsett, W.E. , M . Plocher, V . Wi ldman, D.T. Tingey and J.P. Bennet, 1985. Growth response of two varieties of slash pine seedlings to chronic ozone exposures. Canadian Journal of Botany 63: 2369-2376. Hogsett, W.E. , A . Herstrom, J.A. Laurence, J.E. Weber, E .H. Lee, and D. Tingey, 1995. An approach for characterizing tropospheric ozone risk to forests. In Proceedings of of the 4"" U.S./Dutch International Symposium: Comparative Risk Analysis and Priority Setting for A i r Pollution Issues. Edited by S.D. Lee and T . Scheider. A i r and Waste Management Association, Pittsburgh, pp.521-535. Horsfall, J .G. and R.W. Barratt, 1945. A n improved grading system for measuring plant disease. Phytopathology 35: 655. Humphreys, N. 2001. Personal Communication. Innes, J.L.. 2001. Personal Communication. Innes, J.L. and J .M . Skelly, 1996. Effects of ambient concentrations of ozone on visible symptom development of Prunus serotiha in southern Switzerland. Research Report to the Swiss Federal Institute for Forest, Snow, and Landscape Research, Birmensdorf, Switzerland. Innes, J.L., J .M . Skelly and M . Schaub, 2001. Ozone and broadleaved species: A guide to identification of ozone-induced foliar injury. Haupt (pub), Switzerland. 136pp. Jacob, D J . , J.A. Logan, G .M . Gardner, R.M. Yevich, C M . Spivakovsky, and S.C. Wofsky, 1993. Factors Regulating Ozone Over the United States and Its Export to the Global Atmosphere. Journal of Geophysical Research 98 (D8):14 817- 14 826. Karlsson, P.E., J . Tuovinen, D. Simpson, T . Mikkelsen, and H . Ro-Poulsen, 2002. Ozone Exposure Indices for ICP-Forest Observation Plots within the Nordic Countries. WL Report B1498, I V L Swedish Environmental Institute, Goteberg: 51pp. Kelly N.A., G.T. Wolff, and M.A. Ferman, 1984. Sources and sinks of ozone in rural areas. Atmospheric Environment IS: 1251-1266. Kerstiens, G . and K. J . Lendzian, 1989. Interactions between ozone and plant cuticles. I. Ozone deposition and permeability. New Phytologist 112: 13-19. Kickert, R.N. and S.V. Krupa, 1990. Forest Responses to Tropospheric Ozone and Global Climate Change: A n Analysis. Environmental Pollution 68: 29-65. 153 Kickert, R.N. and S.V. Krupa, 1991. Modelling Plant Response to Tropospheric Ozone: A Critical Review. Environmental Pollution 70: 271-369. Klemm, O., I.C. Ziomass, D. Baylis, P. Suppan, J . Slemr, R. Romero, and L.G. Vyras, 1998. A summer air pollution study in Athens Greece, Atmospheric Environment 32: 2071-2087. Kleinman, L., Y. Lee, S.R. Springston, L. Nunnermacker, X . Zhou, R. Brown, K. Hallock, P. Koltz, D. Leahy, J .H . Lee, and L. Newman, 1994. Ozone formation at a rural site in the southeastern United States. Journal of Geophysical Research 99 (D2): 3469-3482. Krupa, S.V. and W J . Manning, 1988. Atmospheric Ozone: Formation and Effects on Vegetation. Environmental Pollution 50: 101-137. Krupa, S.V. and R.N. Kickert, 1997. Considerations for establishing relationships between ambient ozone (Os) and adverse crop response. Environmental Review 5: 55-11. Krupa, S., M . Nosal, J.S. Ferdinand, R.E. Stevenson, and J .M . Skelly, 2002. A multi-variate statistical model integrating passive sampler and meteorology data to predict the frequency distributions of hourly ambient ozone (Os) concentrations. Environmental Pollution 124:173-178. Lalas, D.P., M . Tombrou-Tsella, M . Petrakis, D.N. Asimakopoulos, and C. Helmis, 1987. A n experimental study of the horizontal and vertical distribution of ozone in Athens, Atmospheric Environment 21: 2681-2693. Lamaud, E., A . Carrara, Y. Brunet, A . Lopez, and A . Druilhet, 2002. Ozone fluxes above and within a pine forest canopy in dry and wet conditions. Atmospheric Environment 36: 77-88. Langebartels, C , W . Heller, G . Fuhrer, M . Lippert, S. Simons, and H . Sandermann Jr., 1998. Memory Effects in the Action of Ozone in Conifers. Ecotoxicology and Environmental Safety 41: 62-72. Lefohn, A.S. 1992. Surface Level Ozone Exposures and their Effects on Vegetation. Lewis Publishers, Chelsea, Michigan, 366pp. Lefohn, A.S. and V .C . Runeckles, 1987. Establishing standards to protect vegetation - ozone exposure/dose considerations. Atmospheric Environment 21(3): 561-568. Lehning, M., H . Richner, G.L. Kok, and B. Neininger, 1998. Vertical exchange and regional budgets of air pollutants over densely populated areas. Atmospheric Environment 32: 1353-1363. Lyons, T .M . , J .D. Barnes and A . W . Davison, 1997. Reltionships between ozone resistance and climate in European populations of Plantago major. New Phytologist 136: 503-510. 154 Manning, W.J., B. Godzik, and R. Musselman, 2002. Potential boindicator plant species for ambient ozone in forested mountain areas of central Europe. Environmental Pollution 119:283-290. McKendry, I.G. 1994. Synoptic Circulation and Summertime Ground-Level Ozone Concentrations at Vancouver, British Columbia. Journal of Applied Meteorology 33(5): 629-641. McKendry, I.G., D.G. Steyn, J . Lundgren, R.M. Hoff, W . Strapp, K. Anlauf, F. Froude, J.B. Martin, R.M. Banta, and L.D. Oliver, 1997. Elevated ozone layers and vertical down-mixing over the Lower Fraser Valley B.C. Atmospheric Environment 31: 2135-2146. McKendry, I.G., D.G. Steyn, R.M. Banta, W . Strapp, K. Anlauf, and J . Potier, 1998. Daytime Photochemical Pollutant Transport over a Tributary Valley Lake in Southwestern British Columbia. Journal of Applied Meteorology 37: 393-404. McKendry, I.G., and J . Lundgren. 2000. Tropospheric layering of ozone in regions of urbanised complex and/or coastal terrain - A review. Progress in Physical Geography 24: 359-384. McLaughlin, D., 1998. A decade of forest tree monitoring in Canada: evidence of air pollution effects. Environmental Review 6: 151-171. McLaughlin, S.B. and D J . Downing, 1995. Interactive effects of ambient ozone and climate measured on growth of mature forest trees. Nature 374: 254-254. Mil lan, M.M., R. Salvador, E. Mantilla, and G. Kallos, 1997. Photo-oxidant dynamics in the western Mediterranean in Summer: Results from European Research Projects. Journal of Geophysical Research 102(D7): 8811-8823. Miller, P.R., G.L. Longbotham, and C.R. Longbotham, 1983. Sensitivity of Selected Western Conifers to Ozone. Plant Disease 67(10): 1113-1115. Miller, P.R. and M J . Arbaugh, 2000. Ozone Impacts on Californian Forests. In: Innes, J.L., and J . Oleksyn (eds.) Forest dynamics in heavily polluted regions (report No . l of the I U F R O task force on environmental change): C A B I Publishing in association with The International Union of Forestry Research, pp. 147-163 Musselman, R.C. and T J . Minnick, 2000. Nocturnal stornatal conductance and ambient air quality standards for ozone. Atmospheric Environment 34(5): 719-733. Neu, U.T. Kunzle, and H . Wanner 1994. O n the relation between ozone storage in the residual layer and daily variation in near surface ozone concentration - a case study. Boundary Layer Meteorology 69: 221-247. 155 Novak, K., J .M . Skelly, M . Schaub, N. Krauchi, C. Hug, W . Landolt and P. Bleuler, 2003. Ozone air pollution and foliar injury development on native plants of Switzerland. Environmental Pollution 125: 41-52. Novak, M.D., J.S. Warland, A .L . Orchanski, R. Keder, and S. Green, 2000. W i n d tunnel and field measurements of turbulent flow in forests. Part I: uniformly thinned stands. Boundary-Layer Meteorology 95: 457-495. Oke, T.R. 1987. Boundary Layer Climates. Roudedge: 435pp. Ollinger, S.V., J .D. Aber and P.B. Reich, 1997. Simulating ozone effects on forest productivity: Interactions among leaf-, canopy-, and stand-level processes. Ecological Applications 7: 1237-1251. Olszyk, D .M . and D. T . Tingey, 1986. Joint Action of Os and SO* in Modifying Plant Gas Exchange. Plant Physiology 82: 401-405. Paakkonen, E., J . Vahala, M . Pahjola, T . Holopainen, and L. Karenlampi, 1998. Physiological, stomatal and ultrastructural ozone responses in birch (Betula pendula Roth.). Plant, Cell and Environment 21: 67'1-684. Paakkonen, E., J., S. Paasisalo, T . Holopainen and L. Karenlampi, 1993. Growth and stomatal responses of birch (Betula pendula Roth.) clones to ozone in open-air and chamber fumigations. New Phytologist, 126: 511-515. Padro, J., H .H . Neumann, and G. Den Hartog, 1991. A n investigation of the A D O M dry deposition module using summertime Os measurements above a deciduous forest Atmospheric Environment 25A: 1689-1704. Pearson, M . and T.A. Mansfield, 1994. Effects of exposure to ozone and water stress on the following growing season's growth of beech (Fagus sylvatica L.). New Phytologist 126: 511-515. Pisano, J.T., I.G. McKendry, D.G. Steyn, and D.R. Hastie 1997. Vertical nitrogen dioxide and ozone concentrations measured from a tethered balloon in the Lower Fraser Valley. Atmospheric Environment 31(4) 2071-2078. Pryor, S.C., I.G. McKendry, and D.G. Steyn, 1995. Synoptic-scale meteorological variability and surface ozone concentrations in Vancouver, British Columbia. Journal of Applied Meteorology 34: 1824-1833. Puxbaum, H., K. Gabler, S. Smidt, and F. Glattes, 1991. A one-year record of ozone profiles in an alpine valley (Zillertal / Tyrol, Austria, 600-2000 m.a.s.1.). Atmospheric Environment 25 A: 1759-1765. 156 Pye, J .M., 1988. Impact of ozone on the growth and yield of trees: a review. Journal of Environmental Quality 17: 347-360. Ray, J.D., and M . Flores, 2000. Evaluation of Passive Samplers for Field Measurements of Ambient Ozone in National Parks. Nature Net, U.S. National Parks Service: http:/Avww2.namre.nps.gov/ard7gas/awmapapr.pdf Reich, P.B. 1983. Effects of Low Concentrations of Os on Net Photosynthesis, Dark Respiration, and Chlorophyll Contents in Aging Hybrid Poplar Leaves. Plant Physiology 1'3: 291-296. Reich, P.B. and R.G. Amundson, 1985. Ambient Levels of Ozone Reduce Net Photosynthesis in Tree and Crop Species, Science 230: 566-570. Reiter, R. 1991. O n the mean daily and seasonal variations of the vertical ozone profiles in the lower troposphere. Atmospheric Environment 25: 1751-1757. Rodon A., C. Johansson, and L. Granat, 1993. Dry deposition of nitrogen dioxide and ozone to coniferous forests. Journal of Geophysical Research 98: 5159-5172. Roper, T.R. and E. Wil l iams, 1989. Effects of Ambient and Acute Partial Pressures of Ozone on Leaf Net CO2 Assimilation of Field-frown Vitis vinifera L. Plant Physiology 91: 1501-1506. Rupprecht & Patashnick Co. Inc., 2002. http.y/www.rpco.corn/products/ambprod/amb3300/ brochure3300_pages.pdf. Salardinoa, D .H . and J J . Carroll, 1998. Correlation between ozone exposure and visible foliar injury in ponderosa and Jeffrey pines. Atmospheric Environment 32(17): 3001-3010. Sahnond, J.A. and I.G. McKendry, 2002. Secondary ozone maxima in a very stable nocturnal boundary layer: observations from the Lower Fraser Valley, BC . Atmospheric Environment 36: 5771-5782. Samuelson, L.J., 1994. Ozone-exposure responses of black cherry and red maple seedlings. Environmental and Experimental Botany 34(4): 355-362. Sandermann, H., A . R. Wel lbum, R.L. Heath, 1997. Forest Decline and Ozone. A Comparison of Controlled Chamber and Field Experiments. Springer, Berlin. Sandroni, S., P. Bacci, G. Boffa, U . Pellegrini, and A . Ventura, 1994. Tropospheric ozone in pre-alpine and alpine regions. The Science of the Total Environment 156: 169-182. Sanz, M J . and M .M . Mil lan, 2000. Ozone in the Mediterranean Region: Evidence of Injury to Vegetation. . In: Innes, J.L., and J . Oleksyn (eds.) Forest dynamics in heavily polluted regions (report No . l of the I U F R O task force on environmental change): 157 C A B I Publishing in association with The International Union of Forestry Research, pp. 165-192. Sather, M.E., J.L. Varns, J .D. Mulik, G. Glen, L. Smith, and C. Stallings, 2001. Passive Ozone Network of Dallas: A Modell ing Opportunity with Community Involvement 2. Environmental Science and Technology? 35: 4426-4435. Senser, M., M . Kloos, and C. Liitz, 1990. Influence of Soil Substrate and Ozone plus Ac id Mist on the Pigment Content and Composition of Needles from Young Spruce Trees. Environmental Pollution 64: 295-312. Sigler, J .M., J .D. Fuentes, R.C. Heitz, M . Garstang, and G . Fisch, 2002. Ozone Dynamics and Deposition Processes at a Deforested Site in the Amazon Basin. Ambio 31(1): 21-27. Sillman, S., J.A. Logan and S.C. Wolfy, 1990. The sensitivity of ozone to nitrogen oxides and hydrocarbons in regional ozone episodes. Journal of Geophysical Research 95(D2): 1837-1851. Sillman, S. 1993. Tropospheric Ozone: The Debate over Control Strategies. Annual Review of Energy and the Environment 18: 31-56. Schaub, M., J .M . Skelly, K.C. Steiner, D.D. Davis, S.P. Pennypacker, J . Zhang, J.A. Ferdinand, J.E. Savage, and R.E. Stevenson, 2003. Physiological and foliar injury responses of Prunus serotina, Fraxinus americana, and Acer rubrum seedlings to varying soil moisture and ozone. Environmental Pollution 124: 307-320. Skelly, J .M., D.D. Davis, W . Merril l, E.A. Cameron, H.D. Brown, D.B. Drummond, and L.S. Dochinger (eds.), 1987. Diagnosing injury to eastern forest trees. USDA-Forest Service, Vegetation Survey Res. Coop, and Pennsylvania State University, University Park, PA . 122 pp. Skelly, J .M., J.E. Savage, M . de Bauer, and D. Alvarez, 1997. Observations of ozone-induced foliar injury on black cherry within the Desierto de Los Leones National Park, Mexico. Environmental Pollution 95: 155-158. Skelly, J .M., J.L. Innes, K.R. Snyder, J.E. Savage, C. Hug, W . Landolt, and P. Bleuer, 1998. Investigations of Ozone induced injury in forests of Southern Switzerland: field surveys and open-top chamber experiments. Chemosphere 36(4-5): 995-1000. Skelly, J .M., J.L. Innes, J.E. Savage, K.R. Snyder, D J . VanderHeyden, J . Zhang, and M J . Sanz, 1999. Observation and confirmation of foliar ozone symptoms of native plant species of Switzerland and southern Spain. Water, Air and Soil Pollution 116: 227-234.. 158 Steyn, D.G. and McKendry, 1988. Quantitative and qualitative evaluation of a three-dimensional mesoscale numerical model simulation of a sea breeze in complex terrain. Monthly Weather Review 116: 1914-1926. Stockwell, W.R:, G. Kramm, H.E. Sheel, V.A. Mohnen and W . Seiler, 1997. Ozone formation, destruction, and exposure in Europe and the United States. In Forest Decline and Ozone A Comparison of Controlled Chamber and Field Experiments. Ecological Studies 127. H . Sanderman, A.R. Wel lburn and R.L. Heath (eds.). Springer-Verlag Heildelberg. Stone, J . 2001. Information Forestry, Canadian Forest Service, Pacific Forestry Centre, Victoria, British Columbia. Natural Resources Canada, August 2001. Stull, R.B. 2000. Meteorology for Scientists and Engineers, T Edition. Brooks/Cole: 502pp. Sturman, A.P. 1987. Thermal influences on airflow in mountainous terrain, Progress in Physical Geography 11: 183-206. Takemoto, B.K., A . Bytnerowicz, M.E. Fenn, 2001. Current and future effects of ozone and atmospheric nitrogen deposition on California's mixed conifer forests. Forest Ecology and Management 144: 159-173. U B C Forestry 2001. University of British Columbia Department of Forest Science, Website. htrp://www.forestry.ubc.ca/refor/general/loceco.htm Unger, L. 2001. Personnal Communication. Van Ooy, D J . , and J . J . Carroll, 1995. The spatial variation of ozone cllimatology on the western slope of the Sierra Nevada. Atmospheric Environment 29(11): 1319-1330. VanderHeyden, D., J . Skelly, J . Innes,'C. Hug, J . Zhang, W . Landolt and P. Bleuler, 2001. Ozone exposure thresholds and foliar injury on forest plants in Switzerland. Environmental Pollution 111: 321-331. Viney, P.A., C.S. Claiborn, Z. L i and A . Murthy, 1994. Trends, seasonal variations and analysis of high-elevation surface nitric acid, ozone and hydrogen peroxide. Atmospheric Environment 28(10): 1781-1790. Wang, D., and F. Bormann, 1986. Regional Tree Growth Reductions due to Ambient Ozone: Evidence from Field Experiments. Environmental Science and Technology 20 (11): 1122-1125. Wanner, H., T . Kunzle, U . Neu, B. Ilhy, G. Baumbach, and B. Steisslinger. 1993. O n die dynamics of photochemical smog over the Swiss Middleland - results in the first P O L L U M E T field experiment Meteorological and Atnospheric Research 15: 117-138. 159 Wesely M.L.. J.A. Eastman, D .H . Stedman, and E.D. Yalvac, 1982.. A n eddy correlation measurement of NO2 flux to vegetation and comparison to Os flux. Atmospheric Environment 16: 815-820. , Woodman, J .N . 1987. Pollution-induced injury in North American forests: facts and suspicion. Tree Physiology 3: 1-15. Woodman, J .N . and E.B. Cowling, 1987. Airborne chemicals and forest health. Environmental Science and Technology 21(2): 120-126. Zaveri, R.A., R.D. Saylor, L.K. Peters, R. McNider, and A . Song 1995. A model investigation of summertime diurnal ozone behaviour in rural mountainous locations. Atmospheric Environment 29(9): 1043-1065. 160 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0091113/manifest

Comment

Related Items