UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of ozone on horticultural crops important in the Fraser Valley of British Columbia Wright, Elaine Frances 1988

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

Item Metadata

Download

Media
831-UBC_1988_A6_7 W74.pdf [ 9.76MB ]
Metadata
JSON: 831-1.0097856.json
JSON-LD: 831-1.0097856-ld.json
RDF/XML (Pretty): 831-1.0097856-rdf.xml
RDF/JSON: 831-1.0097856-rdf.json
Turtle: 831-1.0097856-turtle.txt
N-Triples: 831-1.0097856-rdf-ntriples.txt
Original Record: 831-1.0097856-source.json
Full Text
831-1.0097856-fulltext.txt
Citation
831-1.0097856.ris

Full Text

THE EFFECT OF OZONE ON HORTICULTURAL CROPS IMPORTANT IN THE FRASER VALLEY OF BRITISH COLUMBIA by ELAINE FRANCES WRIGHT B.Sc, Un i v e r s i t y of Toronto, 1982 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Plant Science) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA January 1988 (2) Elaine Frances Wright, 1988 In presenting this thesis in partial fulfilment o f the requirements for an advanced degree at the University o f British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying o f this thesis for scholarly purposes may be granted by the head o f my department or by his or her representatives. It is understood that copying or publication o f this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 DE-6(3/81) ABSTRACT An analysis of a i r q u a l i t y data from B r i t i s h Columbia has i d e n t i f i e d the Lower Mainland and surrounding r u r a l areas as one of the regions i n Canada where the Canadian Maximum Acceptable A i r Quality Objective of 0.082 ppm ozone for one hour i s frequently exceeded. Ozone at t h i s l e v e l has the p o t e n t i a l f o r a f f e c t i n g crops i n the Fraser Valley. F i e l d experiments were undertaken to attempt to evaluate the e f f e c t of randomly f l u c t u a t i n g l e v e l s of ozone on the y i e l d of two c u l t i v a r s each of Brassica oleracea L. ( b r o c c o l i ) , Phaseolus v u l g a r i s L. (bean), Pisum  sativum L. (pea), Daucus carota L. (carrot) i n 1985 and on one c u l t i v a r each of Solanum tuberosum L. (potato) and Pisum sativum L. (pea) i n 1986, using a zonal a i r p o l l u t i o n system. As there i s no current consensus regarding the most appropriate numerical expression of po l l u t a n t exposure to use i n vegetation response studies, a comparison of various exposure terms was a l s o undertaken. Ozone was added i n various proportions to ambient l e v e l s between 0700 and 2100 hours (PDT) throughout the growing season. Three l e v e l s of ozone addi t i o n were used i n 1985 and 12 i n 1986. In 1985, treatments were assigned to three blocks over which ozone was released. Each block was supplied with d i f f e r e n t t o t a l amounts of supplementary ozone, a fourth block serving as an ambient a i r c o n t r o l . In 1986, ozone treatments were randomly assigned to four sub-plots on each of the three blocks oyer which ozone was released, with each block r e c e i v i n g the same t o t a l amount of supplementary ozone. D i f f e r e n t treatments were achieved by each sub-plot being subject to d i f f e r e n t rates of release and degrees of mixing. i i i For both years the ozone concentration d i s t r i b u t i o n s achieved over the season were approximately log-normal. Additional analyis of the a i r q u a l i t y data from the ambient a i r p l o t found other types of skewed d i s t r i b u t i o n s such as the three parameter Weibull, three parameter gamma and Johnsons S B (four parameter log-normal) provided better descriptions of the data. The d i s t r i b u t i o n providing the best f i t depended on the concentration averaging time, the d a i l y time span over which the ozone concentrations are analyzed and the s e l e c t i o n c r i t e r i o n used. In 1985, f i e l d observations indicated that there were numerous p l o t to p l o t d i f f e r e n c e s for disease and s o i l f a c t o r s , which were confounded with the ozone treatments applied. Without true r e p l i c a t i o n of the treatments, d i f f e r e n t i a t i o n between the e f f e c t s due to ozone and those from a b i o t i c and b i o t i c causes was not possible, and hence no cl e a r conclusions concerning ozone response could be drawn. In 1986, without the confounding of ozone and p l o t l o c a t i o n , s i g n i f i c a n t l i n e a r reductions i n y i e l d were found f o r pea and pod fresh weight using the number of days on which the concentration exceeded 25 ppb, during the vegetative growing period (D25 2). A s i g n i f i c a n t l i n e a r reduction i n fresh potato tuber weight was found using the geometric mean of a l l geometric mean ozone concentrations computed between 1200 and 1259-h for the season (GH12) as the exposure s t a t i s t i c . A s i g n i f i c a n t multiple l i n e a r regression was found f o r pea fresh weight using the D25 1, s t a t i s t i c together with the number of occurrences i n which the concentration exceeded 25 ppb for two, three and four consecutive hours (2C25, 3C25 and 4C25 respe c t i v e l y ) i n an episode; and for pod fresh weight using D25, 2C25 and 3C25 as independent v a r i a t e s . i v From the r e s u l t s presented i t seems clear that ozone at the concentrations dispensed here would have a s i g n i f i c a n t negative impact on crops grown i n the Fraser V a l l e y . Based on the 1986 experiment the best case estimate indicates that y i e l d reductions of 28% could be expected for peas and potatoes at 37 ppb ozone (expressed as the season-long 7-h mean, M7). The exposure s t a t i s t i c s used i n the present study i n comparison with the season-long 7 and 12 hour means provided good f i t s with the data. They are e a s i l y c a l c u l a t e d from ambient a i r q u a l i t y data and present a t t r a c t i v e a l t e r n a t i v e s to those exposure s t a t i s t i c s c u r r e n t l y i n use, for assessing the p o t e n t i a l impact of ozone on crops i n the Fraser V a l l e y and for use i n the s e t t i n g of a i r q u a l i t y standards. V TABLE OF CONTENTS ABSTRACT i i TABLE OF CONTENTS V LIST OF TABLES v i i LIST OF FIGURES . i x LIST OF ABBREVIATIONS x i i i ACKNOWLEDGEMENTS xiv 1. INTRODUCTION 1 2 . LITERATURE REVIEW . . 4 2.1 AIR QUALITY 4 2 .2 AIR POLLUTANT DOSE AND EXPOSURE .. 8 2.3 EXPOSURE STATISTICS 13 2.4 EXPOSURE-RESPONSE RELATIONSHIPS 19 2.5 EFFECTS OF OZONE ON CROPS 22 2.6 ENVIRONMENTAL EFFECTS 27 2.7 EXPOSURE METHODOLOGY ..... 31 3. MATERIALS AND METHODS-1985 EXPERIMENT 35 3.1 EXPERIMENTAL DESIGN 35 3.2 FIELD PLOTS AND GAS DELIVERY SYSTEM 35 3.3 OZONE MONITORING AND CONTROL 38 3.4 DATA HANDLING 39 3.5 FIELD PREPARATION 40 3.6 PLANT MATERIALS 40 3.7 CROP AND PEST MANAGEMENT 41 3.8 HARVEST SCHEDULE AND PLANT MEASUREMENTS 41 3.9 DATA ANALYSIS 42 4. RESULTS-1985 EXPERIMENT 45 4.1 AIR QUALITY 45 .1 SPATIAL DISTRIBUTION 45 .2 TEMPORAL DISTRIBUTION 45 4.2 VISUAL OBSERVATIONS 52 4.3 ANALYSIS OF VARIANCE ..55 4.4 REGRESSION ANALYSES 57 .1 SIMPLE LINEAR REGRESSIONS 57 .2 POLYNOMIAL REGRESSIONS 71 5. MATERIALS AND METHODS-1986 EXPERIMENT 76 5.1 EXPERIMENTAL DESIGN 76 5.2 FIELD PLOTS AND GAS DELIVERY SYSTEM 76 5.3 OZONE MONITORING AND CONTROL .81 5.4 DATA HANDLING 83 5.5 FIELD PREPARATION 86 5.6 PLANT MATERIALS 86 5.7 CROP PEST AND RODENT MANAGEMENT 87 5.8 HARVEST SCHEDULE AND PLANT MEASUREMENTS 89 .1 PEA 89 .2 POTATO 89 5.9 DATA ANALYSIS 89 .1 OZONE DISTRIBUTION 89 .2 EXPOSURE STATISTICS 89 .3 REGRESSION ANALYSES 90 6. RESULTS-1986 EXPERIMENT 92 6.1 AIR QUALITY 92 .1 SPATIAL DISTRIBUTIONS 92 .2 TEMPORAL DISTRIBUTIONS 96 6.2 EXPOSURE STATISTICS 121 6.3 REGRESSION ANALYSES 122 . 1 PEA 122 .1 SIMPLE LINEAR REGRESSION ANALYSES 122 .2 POLYNOMIAL REGRESSION ANALYSES 137 .3 STEPWISE MULTIPLE LINEAR REGRESSION ANALYSES ...144 .2 POTATO 149 .1 SIMPLE LINEAR REGRESSION ANALYSES 149 .2 POLYNOMIAL REGRESSION ANALYSES 155 .3 STEPWISE MULTIPLE LINEAR REGRESSION ANALYSES 155 7. DISCUSSION 159 7.1 BACKGROUND 159 7.2 AIR QUALITY 160 7.3 EFFECTS OF OZONE ON CROPS 164 7.4 EXPOSURE STATISTICS 168 7.5 EXPOSURE-RESPONSE FUNCTIONS 175 8. SUMMARY .182 BIBLIOGRAPHY 184 APPENDIX 1 196 APPENDIX 2 198 APPENDIX 3 201 APPENDIX 4 212 v i i LIST OF TABLES TABLE PAGE TABLE 1: L i s t i n g of crops and f i n a l harvest dates for the 1985 f i e l d experiment — 42 TABLE 2: Exposure s t a t i s t i c s used as independent variables i n l i n e a r and polynomial exposure-response models i n 1985 44 TABLE 3: Summary of ANOVA r e s u l t s for the e f f e c t s of ozone on B r o c c o l i v a r i a b l e s (per plant) 56 TABLE 4: Summary of ANOVA r e s u l t s for the e f f e c t s of ozone on Bean variables (per plant) 58 TABLE 5: Summary of ANOVA r e s u l t s for the e f f e c t s of ozone on Pea var i a b l e s (per plant) 59 TABLE 6: Summary of ANOVA r e s u l t s for the e f f e c t s of ozone on Carrot v a r i a b l e s (per p l a n t ) . . . 60 TABLE 7: Results for l i n e a r regressions of B r o c c o l i y i e l d on varous exposure s t a t i s t i c s 61 TABLE 8: Results for l i n e a r regressions of Bean y i e l d on various exposure s t a t i s t i c s 62 TABLE 9: Results for l i n e a r regressions of Pea y i e l d on various exposure s t a t i s t i c s 63 TABLE 10: Results for l i n e a r regressions of Carrot y i e l d on various exposure s t a t i s t i c s 64 TABLE 11: Results for the polynomial regressions of Br o c c o l i y i e l d on various exposure s t a t i s t i c s 72 TABLE 12: Results for the polynomial regressions of Bean y i e l d on various exposure s t a t i s t i c s 73 TABLE 13: Results for the polynomial regressions of Pea y i e l d on various exposure s t a t i s t i c s . . . 74 TABLE 14: Results for the polynomial regressions of Carrot y i e l d on various exposure s t a t i s t i c s 75 TABLE 15: Crops and harvest dates for the 1986 season 87 v i i i TABLE 16: Exposure s t a t i s t i c s used as independent variables i n l i n e a r and polynomial exposure-response models. Except where noted, a l l exposure s t a t i s t i c s r e f e r to values obtained during the d a i l y 12-h period i n which ozone was released, 0900 to 2059 h 91 TABLE 17: Horizontal d i s t r i b u t i o n of ozone i n two locations of the d i s t r i b u t i o n manifold 95 TABLE 18: Simple Linear regressions of pea cv. Puget pod y i e l d s on various exposure s t a t i s t i c s 135 TABLE 19: Simple Linear regressions of pea cv. Puget pea y i e l d s on various exposure s t a t i s t i c s 136 TABLE 20: Polynomial regressions of pea cv. Puget pod y i e l d s on various exposure s t a t i s t i c s 142 TABLE 21: Polynomial regressions of pea cv. Puget pea y i e l d s on various exposure s t a t i s t i c s 143 TABLE 22: Simple Linear regressions of potato cv. Russet Burbank tuber y i e l d s on various exposure s t a t i s t i c s 154 TABLE 23: Polynomial regressions of potato cv. Russet Burbank tuber y i e l d s on various exposure s t a t i s t i c s 158 TABLE 24: Overa l l combined ranking of the exposure s t a t i s t i c s used i n the l i n e a r regression models for a l l crops i n 1986 176 i x LIST OF FIGURES FIGURE PAGE FIGURE 1: Cross-section of a d i c o t leaf showing a series of resistances to p o l l u t a n t gas t r a n s f e r . r a = atmospheric resistance, r s = stomatal resistance and r m = mesophyllic resistance otherwise known as r e s i d u a l resistance. Adapted from O'Dell et a l . , (1977) 11 FIGURE 2: Layout of treatment blocks and network of a d i s t r i b u t i o n manifold of the ZAPS at U.B.C. AA = ambient a i r ; AAxl.5 = ambient a i r x 1.5; AAX2.0 = ambient a i r x 2.0; AAX2.5 = ambient a i r x 2.5. 37 FIGURE 3: V e r t i c a l p r o f i l e of mean ozone concentrations averaged over 2 locations (+ standard de v i a t i o n ) , measured i n the AA X 2.5 p l o t . — > - manifold height (cm).... 47 FIGURE 4: D i s t r i b u t i o n of hourly ozone concentrations (ppb) for 0700 to 2059 hours for AA, AAxl.5, AAx2.0 and AAx 2.5 treatments 49 FIGURE 5: Cumulative d i s t r i b u t i o n s of hourly ozone concentrations (ppb) for 0700 to 2059 for AA, AAxl.5, AAx2.0 and AAx 2.5 treatments, p l o t t e d on l o g - p r o b a b i l i t y paper 51 FIGURE 6: Hourly seasonal means for AA, AAxl.5, AAx2.0 and AAx2.5 treatments 54 FIGURE 7: Relationship between y i e l d (g) for B r o c c o l i cv. Emperor and ozone exposure expressed as P1[12] (ppb) 66 FIGURE 8: Relationship between y i e l d (g) for B r o c c o l i cv. SGI and ozone exposure expressed as P1[14] (ppb) 68 FIGURE 9: Relationship between y i e l d (g) for Bean cv. BBL-GV2 and ozone exposure expressed as PI [12 ] (ppb) 70 FIGURE 10: Network of a d i s t r i b u t i o n manifold of the ZAPS at U.B.C 78 FIGURE 11: Layout of f i e l d experimental p l o t s at U.B.C. Numbers in d i c a t e the frequency of discharge o r i f i c e s over the p l o t s 80 X FIGURE 12: Sampling s i t e s for the ho r i z o n t a l p r o f i l e s measured i n block 2 85 FIGURE 13: V e r t i c a l p r o f i l e of mean ozone concentrations averaged over 2 locations (+ standard d e v i a t i o n ) , i n block 2 —>• = manifold height (cm) . 94 FIGURE 14: Hourly seasonal mean ozone concentrations for block 1, treatments 1 to 4 and AA 1 and 2 for the period June 2 to August 18, 1986 98 FIGURE 15: Hourly seasonal mean ozone concentrations for block 2, treatments 1 to 4 and AA 1 and 2 for the period June 2 to August 18, 1986 100 FIGURE 16: Hourly seasonal mean ozone concentrations for block 3, treatments 1 to 4 and AA 1 and 2 for the period June 2 to August 18, 1986 102 FIGURE 17: D i s t r i b u t i o n s of ozone concentrations (2 minute average between 0900 to 2059, June 2 to August 18) for the ambient a i r p l o t s , 1986 106 FIGURE 18: D i s t r i b u t i o n s of ozone concentrations (2 minute average between 0900 to 2059, June 2 to August 18) for block 1, treatments 1 to 4, 1986 108 FIGURE 19: D i s t r i b u t i o n s of ozone concentrations (2 minute average between 0900 to 2059, June 2 to August 18) for block 2, treatments 1 to 4, 1986 110 FIGURE 20: D i s t r i b u t i o n s of ozone concentrations (2 minute average between 0900 to 2059, June 2 to August 18) for block 3, treatments 1 to 4, 1986 112 FIGURE 21: Cumulative d i s t r i b u t i o n s of ozone concentrations (ppb) (0900 to 2059 h) for the ambient a i r p l o t s , block 4, p l o t t e d on l o g - p r o b a b i l i t y paper... 114 FIGURE 22: Cumulative d i s t r i b u t i o n s of ozone concentrations (ppb) (0900 to 2059 h) f o r treatments 1 to 4, block 1, p l o t t e d on l o g - p r o b a b i l i t y paper.... 116 FIGURE 23: Cumulative d i s t r i b u t i o n s of ozone concentrations (ppb) (0900 to 2059 h) f o r treatments 1 to 4, block 2, p l o t t e d on l o g - p r o b a b i l i t y paper 118 FIGURE 24: Cumulative d i s t r i b u t i o n s of ozone concentrations (ppb) (0900 to 2059 h) for treatments 1 to 4, block 3, p l o t t e d on l o g - p r o b a b i l i t y paper.... 120 x i FIGURE 25: Relationship between the seasonal mean exposure s t a t i s t i c , M7, and Ml[7], P7, and P l [ 7 ] . Data are expressed as r a t i o s of Ml[7], Pl[7] and P7 to M7. The r a t i o s are computed for J u l i a n days 153-211 for pea and 153-228 or 229 for potato .124 FIGURE 26: Relationship between the seasonal mean exposure s t a t i s t i c , M12, and M1[12], P12, and P l [ l 2 ] . Data are expressed as r a t i o s of M l [ l 2 ] , P12 and Pl[12] to M12. The r a t i o s are computed for J u l i a n days 153-211 for pea and 153-228 or 229 for potato ,126 FIGURE 27: Relationship between pod fresh weight (g) of peas cv. Puget and ozone exposure expressed as M7 (ppb). The p r e d i c t i o n l i n e was calculated from the regression equation Y = a + b(M7). = 95% confidence l i m i t s .128 FIGURE 28: Relationship between pea fresh weight (g) of peas cv. Puget and ozone exposure expressed as M7 (ppb). The p r e d i c t i o n l i n e was calculated from the regression equation Y = a + b(M7). = 95% confidence l i m i t s .130 FIGURE 29: Relationship between pod fresh weight-(g) of peas cv. Puget and ozone exposure expressed as M12 (ppb). The p r e d i c t i o n l i n e was calculated from the regression equation Y = a + b(M12). = 95% confidence l i m i t s .132 FIGURE 30: Relationship between pea fresh weight (g) of peas cv. Puget and ozone exposure expressed as M12 (ppb). The p r e d i c t i o n l i n e was calculated from the regression equation Y = a + b(M12). = 95% confidence l i m i t s .134 FIGURE 31: Relationship between pod fresh weight (g) of peas cv. Puget and ozone exposure expressed as D25 2 (153-179) (ppb). The p r e d i c t i o n l i n e was calc u l a t e d from the regression equation: Y = a + b(D25 2). = 95% confidence l i m i t s 139 FIGURE 32: Relationship between pea fresh weight (g) of peas cv. Puget and ozone exposure expressed as D25 2 (153-179) (ppb). The p r e d i c t i o n l i n e was calc u l a t e d from the regression equation: Y = a + b(D25 2). = 95% confidence l i m i t s .141 x i i FIGURE 33: Relationship between pea cv. Puget, pod fr e s h weight (g) and ozone exposure expressed as D25 2 (153-179). The p r e d i c t i o n l i n e was calculated from the regression equation: Y = a + b!(D25 2) + b 2 ( D 2 5 2 ) 2 . = 95% confidence l i m i t s 146 FIGURE 34: Relationship between pea cv. Puget, pea fresh weight (g) and ozone exposure expressed as D25 1 (153-211). The p r e d i c t i o n l i n e was calculated from the regression equation: Y = a + b 1(D25 1) + b 2 ( D 2 5 1 ) 2 . = 95% confidence l i m i t s . 148 FIGURE 35: Relationship between potato cv. Russet Burbank tuber fresh weight (g) and ozone exposure expressed as M7 (ppb). The p r e d i c t i o n l i n e was calculated from the regression equation: Y = a + b(M7). = 95% confidence l i m i t s 151 FIGURE 36: Relationship between potato cv. Russet Burbank tuber fresh weight (g) and ozone exposure expressed as M12 (ppb). The p r e d i c t i o n l i n e was calculated from the regression equation: Y = a + b(M12). = 95% confidence l i m i t s 153 FIGURE 37: Relationship between potato cv. Russet Burbank tuber fresh weight (g) and ozone exposure expressed as GH12 (ppb). The p r e d i c t i o n l i n e was calculated from the regression equation: Y = a + b(GH12). = 95% confidence l i m i t s 157 x i i i ABBREVIATIONS ANOVA analysis of variance CERL C o r v a l l i s Environmental Research Laboratory CF charcoal f i l t e r e d a i r cm centimetre h hour m metre min minute NCLAN National Crop Loss Assessment Network NF n o n - f i l t e r e d a i r O3 ozone ppb parts per b i l l i o n (v/v) ppm parts per m i l l i o n (v/v) r a atmospheric resistance r s stomatal resistance r r r e s i d u a l resistance ZAPS zonal a i r p o l l u t i o n system x i v Sincere thanks to the members of my committee, i n p a r t i c u l a r my supervisor Dr. V.C. Runeckles for the many hours spent e d i t i n g my thesis and h i s guidance and encouragement throughout t h i s program. I would also l i k e to thank Loretta M i k i t z e l , Kathy Penney, Chris Pike, P h i l i p Ross, and the Geophysics gang f or t h e i r assistance i n c o l l e c t i n g data. Many thanks to Derek White and Peter Garnett for t h e i r t e c h n i c a l assistance i n running the f i e l d research, without which i t would not have taken place. To C h r i s t i a Roberts, many thanks for the advice on growing crops. I am gr a t e f u l to Dr. G.W. Eaton f o r h i s advice and willingness to spend time a s s i s t i n g me with the s t a t i s t i c a l a n a l y s i s . F i n a l l y , I extend thanks to my husband P h i l i p , whose enthusiastic support throughout t h i s program was g r a t e f u l l y appreciated. 1 1. INTRODUCTION Ozone, the dominant component of photochemical a i r p o l l u t i o n i s a common problem i n large urban centres. A report by Wilson et a l . i n 1984 i d e n t i f i e d the Lower Mainland of B r i t i s h Columbia and surrounding r u r a l areas as one of the four worst-affected regions i n Canada, i n which the Canadian Maximum Acceptable A i r Quality Objective of 0.082 ppm ozone for one hour i s frequently exceeded. Though i t has been well documented i n the l i t e r a t u r e that ozone a l t e r s the growth and y i e l d of plants (Jacobson, 1982; Heagle et a l . , 1983; Kress et a l . , 1985), much of the research was based on growth chamber and greenhouse studies which have l i t t l e p r e d i c t i v e power i n e s t a b l i s h i n g the e f f e c t s of ozone on crops growing under f i e l d conditions (Medeiros et a l . , 1983; Jacobson, 1982). This i s because plant responses to a i r po l l u t a n t s are varied, being influenced by c u l t i v a r s e n s i t i v i t y , environmental conditions, the presence of other p o l l u t a n t s (Roberts, 1984; Heggestad et a l . , 1985), pollutant concentration and the number, timing and duration of exposures (Jacobson, 1982). There are a number of approaches cur r e n t l y i n use for assessing the impact of a i r pollutants on crop y i e l d i n f i e l d s i t u a t i o n s . These include the "open-top" chamber, used by the National Crop Loss Assessment Network (NCLAN) (Heck et a l . , 1982), the open-air zonal a i r p o l l u t i o n system (ZAPS) (Lee et a l . , 1975), and the open-air, chamberless approach, where natural p o l l u t a n t gradients are u t i l i z e d (Oshima et a l . , 1976) . The open-top chamber approach suffers from various l i m i t a t i o n s , the most important of which are the modification of the plants environment by the chambers themselves, and the i n a b i l i t y to assess p o l l u t a n t treatment x / 2 chamber i n t e r a c t i o n s . The ZAPS approach involves the release of gaseous pol l u t a n t s above f i e l d p l o t s , leading to a modification of the composition of the ambient a i r without modifying the microclimate around the p l a n t s . S i m i l a r l y , no d e l i b e r a t e modification of the microclimate occurs with the use of natural p o l l u t a n t gradients, but the lack of control of environmental variables i n general l i m i t s i t s a p p l i c a b i l i t y . Furthermore, the use of such gradients i s l i m i t e d to s i t u a t i o n s i n which a i r q u a l i t y i s determined only by a single p o l l u t a n t . Regardless of the experimental approach used, such f i e l d research i s usually aimed at developing pollutant treatment-response r e l a t i o n s h i p s that may ultimately be used to estimate the e f f e c t s of differences i n ambient a i r q u a l i t y on crops and other vegetation. The r e l a t i o n s h i p s take the form of mathematical functions of various types, i n which some measure of growth, y i e l d or y i e l d loss i s r e l a t e d to the l e v e l ( s ) of p o l l u t a n t ( s ) to which the crop has been exposed. One of the problems i n describing exposure response r e l a t i o n s h i p s i s the s e l e c t i o n of the numerical d e f i n i t i o n of the exposure s t a t i s t i c (Krupa et a l . , 1986). At present there i s no consensus on an exposure s t a t i s t i c that best r e l a t e s the p o t e n t i a l response of the plants to the f l u c t u a t i n g ozone concentrations experienced by vegetation i n the f i e l d over a growing season. In view of the ozone problem i n the Lower Mainland and the lack of relevant information concerning the possible e f f e c t s of randomly f l u c t u a t i n g concentrations of ozone t y p i c a l of ambient conditions on vegetable crops grown i n t h i s region, my study was designed to achieve the following objectives. 3 1. To determine the yield responses of important vegetable crops grown in the Fraser Valley (broccoli, beans, peas, carrots, lettuce and potatoes) to randomly fluctuating ozone concentrations under field conditions. 2. To evaluate various numerical expressions of ozone exposure and develop exposure-response functions for each of the crops studied. 4 2. LITERATURE REVIEW 2.1 A i r Quality I n i t i a l l y considered to be an urban p o l l u t a n t , ozone (0 3 ) i s now recognized as a p o l l u t a n t of regional and global s i g n i f i c a n c e (Fowler and Cape, 1982; Jacobson, 1982). It i s a secondary a i r p o l l u t a n t , formed by photochemical reactions in v o l v i n g anthropogenic emissions of nitrogen oxides and hydrocarbons. As the production of nitrogen oxides and hydrocarbons i s p r i m a r i l y linked to d a i l y urban a c t i v i t y , there i s a t y p i c a l d i u r n a l c y c l e with an early afternoon maximum (Luria et a l . , 1984). In Canada and much of the U.S. and Europe, the highest O3 concentrations are usually recorded i n June and July, corresponding to the primary growing period f or crops. Anthropogenic O3 concentrations t y p i c a l l y s t a r t with a buildup of NO and N0 2 and hydrocarbons i n the morning leading to a peak i n O3 concentration i n the early afternoon. The concentration of O3 drops o f f towards l a t e afternoon or evening. Data c o l l e c t e d at a number of r u r a l s i t e s by K e l l y et a l . (1984) suggested that, i n these locations, the d i u r n a l increase i n O3 comprised two f a c t o r s : an e a r l y r i s e (0600 to 1000 hours) due to downmixing of O3 and a subsequent increase (1000 to 1400 hours) due to photochemical reactions of transported precursors and further downmixing. Ozone depletion at night generally occurs between 1800 and 0600 hours (h) while nocturnal inversions prevent mixing with a i r from a l o f t . The depletion i s thought to be due to dry deposition and chemical scavenging reactions, i f the necessary chemical reactants (N0„ and terpenes) are present ( K e l l y et a l . , 1984). However, persistence of 0 3 a f t e r dark i s possible, p a r t i c u l a r l y i n r u r a l areas, where the a i r t y p i c a l l y contains smaller amounts of compounds that scavenge 0? (Jacobson, 5 1982). The occurrence of high O3 concentrations i s therefore dependent on a v a r i e t y of dynamic fa c t o r s such as wind, t r a f f i c , emissions from power plants and other i n d u s t r i a l sources, sunlight and large scale a i r movements. Pratt et a l . (1983) examined data from three r u r a l s i t e s i n the mid-western, U.S. for the years 1977 to 1981 and found that long range transport of O3 and i t s precursors from urban areas and stratospheric O3 i n t r u s i o n s were the primary sources of elevated r u r a l 0 3 l e v e l s . Similar conclusions were reached by several other researchers (Evans et a l . , 1983; K e l l y et al.» 1984; Chung and Dann, 1985; Colbeck and Harrison, 1985). In an analysis of a i r q u a l i t y data covering the period of 1978 to 1981 by Environment Canada, Vancouver ranked t h i r d highest i n Canada for the number of " s t a t i o n episodes" and second highest for number of "episode days", based upon the frequencies of exceedances of the Canadian A i r Quality Objectives for O 3 . Unlike other high O3 areas i n Canada, the Lower Mainland 0 3 precursors r e s u l t from l o c a l urban emissions (Wilson et a l . , 1984). T y p i c a l l y , O3 episodes i n the Lower Mainland are associated with stable weather systems i n the area, providing clear skies and calm winds, conducive to the formation and buildup of O 3 . The data suggest that O3 and i t s precursor pollutants are being transported by daytime sea breezes from the Greater Vancouver Regional D i s t r i c t (GVRD) i n t o the Fraser V a l l e y . For an assessment of the r e l a t i o n s h i p betweeen O3 l e v e l s and crop response, accurate measurement of the p o l l u t a n t and c h a r a c t e r i z a t i o n of i t s concentration d i s t r i b u t i o n are required (Fowler and Cape, 1982). The frequency of occurrence of high concentrations i s of i n t e r e s t both from the point of view of the e f f e c t s on crops and for s e t t i n g a i r q u a l i t y standards. Once the observed concentrations are f i t t e d to a s t a t i s t i c a l 6 d i s t r i b u t i o n , p r e d i c t i o n of the occurrence of extreme values i s p o s s i b l e (Fowler and Cape, 1982). A i r p o l l u t a n t concentrat ions are by nature random v a r i a b l e s (Georgopoulos and S e i n f e l d , 1982) that are frequent ly autocorre la ted and tend to d i s p l a y random f l u c t u a t i o n s due to changes i n meteorologica l condi t ions and p o l l u t a n t emissions (Holland and F i t z -S imons , 1982; Chock and Sluchak, 1986). Therefore the bas ic assumptions of independence and i d e n t i c a l l y d i s t r i b u t e d v a r i a b l e s for a i r p o l l u t i o n data are not v a l i d (Georgopoulos and S e i n f e l d , 1982). Th i s v i o l a t e s the assumptions for goodness of f i t t e s t s (Taylor et a l . , 1986). F u r t h e r , secondary p o l l u t a n t s , un l ike primary p o l l u t a n t s , may not be we l l descr ibed by a log-normal d i s t r i b u t i o n , because t h e i r concentrat ions tend to be dependent on l o c a l i z e d chemical processes as opposed to atmospheric d i s p e r s i o n (Fowler and Cape, 1982; Nosal , 1984). Under these circumstances, a number of d i s t r i b u t i o n s may be necessary to adequately descr ibe the da ta , p a r t i c u l a r l y i f they span a long p e r i o d of time (Holland and F i t z -S imons , 1982). Georgopoulos and S e i n f e l d (1982) and Male (1982) a l so pointed out that concentrat ions averaged over long time per iods tend to be less w e l l c o r r e l a t e d than those averaged over shorter p e r i o d s . Bencala and S e i n f e l d (1976) and Georgopoulos and S e i n f e l d (1982) pointed out that , i n most cases, no one d i s t r i b u t i o n may f i t a l l ranges of p o l l u t a n t concentrat ion i n a data se t . Georgopoulos and S e i n f e l d (1982) fur ther suggested that i t may be appropriate for such data sets to be d i v i d e d i n t o h igh versus low concentrat ion ranges with d i f f e r e n t d i s t r i b u t i o n s f i t t e d to each reduced data se t . Caution has been advised by Curran and Suggs (1986) i n r e l y i n g on high concentrat ions for s e t t i n g a i r q u a l i t y standards, as they are 7 associated with greater uncertainty due to measurement imprecision. The ef f e c t of t h i s bias i s diminished for summary s t a t i s t i c s i n the middle of the d i s t r i b u t i o n . Nosal (1984) found Weibull and gamma functions best described the d i s t r i b u t i o n of ambient O3 concentration data obtained at the University of Minnesota. Larsen (1973), following a graphical analysis of a i r p o l l u t i o n data, found that regardless of the pol l u t a n t , averaging time or c i t y , a i r p o l l u t a n t concentrations were best described by 2-parameter log-normal d i s t r i b u t i o n s . He further found that concentration data obtained near i s o l a t e d point sources were best represented by a 3-parameter log-normal d i s t r i b u t i o n . . This i s i n contrast to the work of Taylor et a l . (1986), who evaluated the f i t of.a number of d i s t r i b u t i o n s to a i r q u a l i t y data (24-h average concentrations) recorded i n Melbourne, A u s t r a l i a . They, l i k e Nosal (1984), found the gamma and Weibull d i s t r i b u t i o n s were appropriate for various carbon monoxide and O3 data sets, but log-normal d i s t r i b u t i o n s were best for p a r t i c u l a t e data and a number of n i t r i c oxide, oxides of nitrogen and sulphur dioxide data sets. Bencala and Se i n f e l d (1976), using 1-h average carbon monoxide data and the sum of squares error as the goodness of f i t c r i t e r i o n , found a 3-parameter log-normal d i s t r i b u t i o n to be best. They suggested that the approximate log-normal d i s t r i b u t i o n could be i n part explained by the instantaneous a i r p o l l u t a n t concentration d i s t r i b u t i o n s . These tended to be approximately log-normal i f the wind speed was log-normally d i s t r i b u t e d . Chock and Sluchak (1986) i n an ana l y s i s of a i r q u a l i t y data u t i l i z i n g the Maxfit package (introduced by Holland and Fitz-Simons, 1982) found that the goodness of f i t t e s t s d i d not c o n s i s t e n t l y provide a clear ranking of the d i s t r i b u t i o n s f i t t e d . A change i n data base changed the goodness of 8 f i t s t a t i s t i c s of each d i s t r i b u t i o n . They found that o u t l i e r s on either extreme of the data set had large uncertainties and were s e n s i t i v e to the d i s t r i b u t i o n f i t t e d . Values that were les s extreme had smaller uncertainties and were less s e n s i t i v e to the form and parameters of the f i t t e d d i s t r i b u t i o n . They concluded (independently of Curran and Suggs, 1986) that use of extreme values i n s e t t i n g a i r q u a l i t y standards i s associated with large uncertainty and recommended that less-extreme values (that tend to be less s e n s i t i v e to autocorrelation) are the more appropriate choice for a i r q u a l i t y standards. Buttazzoni et a l . (1986) also u t i l i z e d the Holland and Fitz-Simons. (1982) Maxfit program to f i t the most appropriate d i s t r i b u t i o n s to S0 2 data from Venice. To improve the homogeneity of the data set they divided i t i n t o sub-data sets by season and time of day (to account for the work day, nocturnal inversions, e t c . ) . The best o v e r a l l f i t was found with a Weibull d i s t r i b u t i o n , though for the summer data the Weibull model tended to over-estimate the 89th p e r c e n t i l e . 2.2 A i r Pollutant Dose and Exposure To create a phytotoxic event, O3 i n the ambient a i r must d i f f u s e i n t o the leaf of the plant. Gaseous d i f f u s i o n through the stomata i s the primary route of O3 entry. Rates of absorption i n t o the plant may be r e l a t e d to external and i n t e r n a l resistances, i n s e r i e s . Models for p o l l u t a n t uptake, i . e . f l u x , by leaves and vegetative surfaces, have been developed by a number of authors (Bennett et a l . , 1973; O'Dell et a l . , 1977; Baldocchi et a l . , 1987). The basis for many of these models i s 9 d i f f u s i o n theory. By analogy to Ohm's law, t h i s can be expressed as: Flux (jug i n - 2 s ~ i ) = Concentration gradient pug n d ) density Resistance (s m _ 1) (1) The gradient i s the concentration d i f f e r e n c e between the a i r and the ultimate sink for the p o l l u t a n t within the c e l l s of the leaf which i s considered to have a concentration of zero (Unsworth, 1982). The t o t a l resistance i n equation (1) can be p a r t i t i o n e d i n t o a s e r i e s of component resistances: leaf boundary layer ( r a ) , stomatal ( r s ) , and r e s i d u a l or mesophyllic resistance ( r r ) . This r e s u l t s i n the following equation: Deposition v e l o c i t y (m s _ 1 ) = l / [ r a + r g + r f ] (2) (Unsworth, 1982). A c r o s s - s e c t i o n a l view of a t y p i c a l d i c o t leaf i s shown i n Figure 1, with the resistances superimposed upon i t . Exposure, dose and f l u x are terms frequently encountered i n the a i r p o l l u t a n t l i t e r a t u r e , and are often used interchangeably. However, Lefohn and Runeckles (1987) have pointed out that exposure r e f e r s to the pollutant concentration i n the ambient a i r (or treatment) monitored at the plant canopy over time, whereas dose r e f e r s to that portion of the pollutant absorbed by the plant, and i s based on the concept of " e f f e c t i v e dose" introduced by Runeckles (1974) to describe that portion of the ambient O3 that i s involved i n causing i n j u r y . The e f f e c t i v e dose estimated by f l u x i s more representative of what acts on the plant than i s the ambient concentration. In a s i m i l a r vein, Fowler and Cape (1982) developed the concept of p o l l u t a n t absorbed dose (PAD), i n units of g m . This i s 10 Figure 1: C r o s s - s e c t i o n of a d i c o t l ea f showing a ser ies of res i s tances to p o l l u t a n t gas t r a n s f e r . r a = atmospheric r e s i s t a n c e , r s = stomatal res is tance , and r m = mesophyl l ic r e s i s t a n c e . Adapted from O ' D e l l et a l . (1977). 11 12 obtained by taking the product of the ambient concentration of the gas concerned, time and stomatal or canopy conductance. This approach i s supported by the work of Macdowall et a l . (1964) and Mukammal (1965) who attempted to r e l a t e the occurrence of weather f l e c k i n j u r y i n tobacco to ambient 0 3 concentrations and found a poor c o r r e l a t i o n u n t i l a c o e f f i c i e n t of evaporation based on evapotranspiration was developed and inserted into the r e l a t i o n s h i p . Once exposure was corrected i n t h i s way, "dose" and i n j u r y were found to be r e l a t e d l i n e a r l y . It was suggested that the c o e f f i c i e n t of evaporation was proportional to evapotranspiration and hence resembled the mechanism of gas exchange. Percent f o l i a r i n j u r y was then dependent on the amount of pollutant absorbed. There i s evidence i n the l i t e r a t u r e that points to the importance of high, short term peaks as opposed to low, long term concentrations i n causing i n j u r y and adverse growth e f f e c t s on a number of h o r t i c u l t u r a l crops (Heck et a l . , 1966; Reinert and Nelson, 1979; Nouchi and Aoki, 1979; Musselman et al.,1983; Amiro et a l . , 1984; Hogsett et a l . , 1985). Heck et a l . , (1966) working with Nicotiana tabacum L. (tobacco) and Phaseolus  v u l g a r i s L. (bean) found a c u r v i l i n e a r r e l a t i o n s h i p between plant response, exposure time and a wide range of O3 concentrations. Peak concentrations were more damaging than lower concentrations amounting to the same t o t a l exposure ( i . e . concentration x time), given over a longer period of time. However the work was based on exposure rather than dose or e f f e c t i v e dose. Recognizing the importance of uptake i n b i o l o g i c a l response, Amiro et a l . (1984) attempted to r e l a t e the onset of v i s i b l e f o l i a r i n j u r y i n bean to 0 3 f l u x density. They found the s e n s i t i v i t y of plants to 0 3 was associated with stomatal conductance. There was a c u r v i l i n e a r r e l a t i o n s h i p between O3 concentration and i n j u r y ; a threshold f l u x density existed below 13 which f o l i a r i n j u r y d i d not occur. They further suggested that f l u x density should replace p o t e n t i a l dose i n exposure-response experiments. Taylor et a l . (1982) investigated the r e l a t i o n s h i p between f o l i a r i n j u r y , p o l l u t a n t f l u x and leaf morphology i n Glycine max L. Merr. cvs. Hood and Dare (soybean). However, they found that stomatal and boundary layer resistance d i d not f u l l y account for v a r i a t i o n i n p o l l u t a n t uptake rates i n t o the leaf i n t e r i o r . Furthermore, r e s i d u a l resistance to O3 f l u x increased with pollutant concentration and exposure time and was associated with age-dependent dif f e r e n c e s i n f o l i a r 0 3 response. The extent of f o l i a r i n j u r y was not c o n s i s t e n t l y r e l a t e d to the magnitude of O3 f l u x i n t o the le a f i n t e r i o r . It was suggested that the gas and l i q u i d phase pathways for O 3 , H2O vapor and C0 2 are not i d e n t i c a l . 2.3 Exposure s t a t i s t i c s A l l exposure-response models are dependent on the method of expressing the exposure. The s e l e c t i o n of an appropriate mathematical expression of exposure (based s o l e l y upon ambient a i r concentrations) s t i l l presents a problem i n the development of r e a l i s t i c exposure-response functions, because of the uncertain r e l a t i o n s h i p between exposure and true dose discussed above. To date there i s no consensus on an exposure s t a t i s t i c ( s ) that w i l l best r e l a t e the response to the plant i n the f i e l d to randomly f l u c t u a t i n g O3 concentrations over a growing season (Heck et a l . , 1984a; Hogsett et a l . , 1985; Krupa and Kickert, 1987; Runeckles 1987). Lefohn and Runeckles (1987) have out l i n e d a number of issues c e n t r a l to the development of an appropriate numerical summary of exposure. These include the following: 1) the s e n s i t i v i t y of the plant at the time of exposure, 2) the amount and form of the po l l u t a n t entering the plant, 3) 14 the length of the episodic event and 4) the time between exposures. Nouchi and Aoki (1979), working with P h a r b i t i s n i l var. Scarlet 0'Hara (Japanese morning glory) addressed the issue of cumulative exposure and found that, as the amount of time between successive exposures increased, the percent f o l i a r i n j u r y decreased. Krupa and Teng (1982) as c i t e d i n Krupa and Kickert (1987) addressed the importance of the developmental stage of the plant i n r e l a t i o n to i t s a b i l i t y to deal with pollutant exposure. They suggested that the b i o l o g i c a l response of the target plant to the pollutant exposure v a r i e d i n r e l a t i o n to the s e n s i t i v i t y of the developmental stage. Exposure s t a t i s t i c s that have been used to date include mean values for 1, 3, 6, 7, 12, or 24 hours , both d a i l y and averaged over a longer period (e.g. a calander quarter, a "growing season" or a year); seasonal maximum hourly mean concentrations; seasonal sums of absolute hourly average concentrations above a threshold (e.g. 0.05, 0.10 or 0.15 ppm); seasonal sums of fr a c t i o n s of hourly concentrations above a threshold (e.g. 0.08, 0.10 ppm); frequency d i s t r i b u t i o n s of hourly mean concentrations and seasonal sums of hours above a threshold (e.g. 0.08, 0.10 ppm), as outlined i n Jacobson (1982). In addition, other exposure s t a t i s t i c s used have included the use of an exponential function to weight O3 concentrations (e.g. C x t, where C = p o l l u t a n t concenration and t = exposure time, Nouchi and Aoki, 1979); and the " e f f e c t i v e mean O3 concentration": M e = [OF. C h ~ 1 / , v ) / N ] - v , where C h = the hourly average ambient 0 3 concentration during the exposure period (0900 to 1559 h), n = the t o t a l number of hours and v = an exposure time-concentration parameter to assign greater weight to the high concentrations (Larsen and Heck, 1984). Larsen and Heck (1984) pointed out that two s i t e s with s i m i l a r arithmetic mean O3 concentrations may have d i f f e r e n t e f f e c t i v e mean O3 concentrations and 15 therefore d i f f e r e n t estimated crop reductions. Once again t h i s emphasizes the problem with long term averaging of pollutant concentrations. Use of any of these exposure s t a t i s t i c s involves c e r t a i n assumptions. C a l c u l a t i o n of the widely used seasonal exposure s t a t i s t i c s such as the M7 or M12 (the seasonal arithmetic means of the d a i l y 7- or 12-h averages for the periods commencing at 0900 h, respectively) are based on the assumption that the data are normally d i s t r i b u t e d . The sum of a l l concentrations during the exposure period (ppm.h), used by Foster et a l . (1983b), tre a t s a l l concentrations as being equally e f f e c t i v e i n a f f e c t i n g crop y i e l d which i s also true of long-term mean values. Neither of these s t a t i s t i c s i s responsive to the period of time i n which the plant i s most suceptible on a day-to-day b a s i s . Those exposure s t a t i s t i c s with a threshold assume that i t i s only the higher concentrations which are important i n determining plant response. Oshima et a l . (1976) chose 0.10 ppm as the threshold l i m i t and used the sum of the f r a c t i o n s of the concentrations above t h i s threshold as the exposure s t a t i s t i c (ppm.h) i n t h e i r study of a l f a l f a , using a natural O3 gradient i n Southern C a l i f o r n i a . This threshold was selected because r e s u l t s were inconclusive with the range of thresholds examined (0.03 to 0.15 ppm) and 0.10 ppm represents the 1-hour maximum C a l i f o r n i a n a i r q u a l i t y standard. Lefohn and Benedict (1982) also made the assumption that the higher concentrations play the dominant r o l e i n determining plant response to O3 They introduced the term "integrated exposure" index (ppm.h), obtained by summing the products of absolute concentration and t o t a l time when the concentration exceeded the threshold of 0.10 ppm. As Lefohn and Runeckles (1987) point out, both of these i n d i c e s f a i l to consider recovery between 16 episodes and diminish the importance of lower concentrations i n a l t e r i n g the plants s u s c e p t i b l i t y to subsequent episodes. Completely ignoring low l e v e l s of O3 i s inappropriate because pretreatment with low l e v e l s of O3 has been found by a number of researchers to change s e n s i t i v i t y to subsequent O3 exposure (Runeckles and Rosen, 1974; Johnston and Heagle, 1982) Increased s e n s i t i v i t y of soybean plants to an acute exposure of 0.2 ppm for 3 hours was found when using a pre-treatment exposure of 0.06 ppm i n comparison to those grown i n c h a r c o a l - f i l t e r e d (CF) a i r (Johnston and Heagle, 1982). Bennett et a l . , (1974), Runeckles and Rosen (1974) and Johnston and Heagle (1982) have pointed out that the use of CF control treatments ( p o l l u t a n t - f r e e c o n t r o l s ) , achieved by rigorous f i l t r a t i o n of ambient a i r may not be not be r e a l i s t i c i n r e l a t i o n to f i e l d conditions and could lead to misleading r e s u l t s . An improvement on the integrated exposure index i s the l o g i s t i c or sigmoidal weighting function proposed by Lefohn and Runeckles (1987). This function weights a l l i n d i v i d u a l concentrations and sums them over time; thus low concentrations which may play an important r o l e i n e l i c i t i n g a response i n the crop are not ignored. The weighting function proposed i s : WL = 1/[1 + M x exp ( - A x CL)] (3) where M and A are constants, and w^  = the weighting f a c t o r f o r concentration, C^. The product of each concentration and weight (C^ x W^ ) replaces the concentration term per se i n computing exposure s t a t i s t i c s . These authors also proposed a further modification that incorporates the concept of duration and time between episodes. The time between exposures 17 i s of i n t e r e s t as i t addresses the issue of p o s s i b l e recovery by the plant between subsequent exposures. The method was o r i g i n a l l y developed by Mancini (1983) for c a l c u l a t i n g m o r t a l i t y of aquatic organisms following time-varying toxicant concentrations. The equation presented by Lefohn and Runeckles (1987) i s as follows: n n Integrated Exposure = V w i x c i + H c t i e x P ^ - k r t-P i^O j=l where = the sigmoidal weighting function at hour i presented i n Equation 3, c^ = ambient hourly average concentration at.hour i , C t j = the l a s t hourly concentration above threshold before episode j ended, k r = d e t o x i f i c a t i o n rate, and t j = time between episode j and j + 1. Although Equation 4 has yet to be tested, there i s now evidence supporting the weighting approach of Equation 3 i t s e l f (Lefohn et a l . , 1988). Nosal (1984) addressed the matter of e p i s o d i c i t y of p o l l u t a n t concentration fluxes i n a i r q u a l i t y data i n a f f e c t i n g plant response i n a d i f f e r e n t way . He attempted to encompass these aspects i n t o the following exposure parameters: X± = the number of p o l l u t a n t episodes (days where the concentration exceeded 1 ppb) X 2 = the i n t e g r a l of concentration over each episode accumulated over the experiment (concentration and duration, pbb.h) Xg = the s i n g l e peak po l l u t a n t concentration during the t o t a l exposure period. 18 In turn, these were then used i n a m u l t i v a r i a t e polynomial Fourier model of plant response (see below, Section 2.4). The large U.S. National Crop Loss Assessment Network (NCLAN) program selected the seasonal 7-h d a i l y mean O3 concentration as the exposure s t a t i s t i c f o r use i n response models because of i t s s i m p l i c i t y i n computation (Heck et a l . , 1984a), and i n s p i t e of i t s underlying assumption that a l l concentrations are equally important i n e l i c i t i n g a response i n the pl a n t . Krupa and Kickert (1987) and Runeckles (1987) have pointed out that season-long averaging assumes that the pollutant concentrations are normally d i s t r i b u t e d whereas work by Lefohn and Benedict (1982) and Nosal (1984) have found that the Weibull function best decribes the ambient a i r concentrations. Krupa and Kickert (1987) suggested that medians and p e r c e n t i l e s which have been calculated from short duration p o l l u t a n t monitoring so that they are free from the non-normal d i s t r i b u t i o n may be more appropriate summaries of the p o l l u t a n t exposure. This had a l s o been suggested by Male (1982). The e f f e c t of inappropriately averaging non-normal data i s i l l u s t r a t e d by the work of Hogsett et a l . (1985), who compared the response of a l f a l f a exposed to episodic and d a i l y peak exposure p r o f i l e s with equivalent integrated exposures over the growing period. When exposure was expressed as a seasonal 7-h mean the episodic p r o f i l e with the smaller seasonal mean caused greater y i e l d reductions. C l e a r l y the 7-h means i n t h i s case d i d not r e f l e c t the nature of the treatment applied. In attempting to rebut the c r i t i c i s m s l e v e l l e d at the use of M7 seasonal means, NCLAN assessed a few a l t e r n a t i v e exposure s t a t i s t i c s for t h e i r use i n exposure response models: the season-long 1-h grand mean (Ml), peak 1-h (Pl) concentration i n the season and peak 7-h (P7) average i n the 19 season (Heck et a l . , 1984b). These s t a t i s t i c s were compared by assessing the v a r i a t i o n i n the r a t i o s of Ml, PI, and P7 to the M7. Poor r e l a t i o n s h i p s were found for the peak r a t i o s , P1/M7 and P7/M7 (Heck et a l . , 1984b). Cure et a l . (1986) found y i e l d p r e dictions were s i m i l a r using ei t h e r the M7 or Ml. However, large differences were found with the P1/M7 r a t i o s among treatments i n the same experiment for d i f f e r e n t years, and y i e l d s predicted with the peak s t a t i s t i c s were 3 to 4 times higher than those using the seasonal means. 2.4 Exposure-Response Relationships The exposure-response concept i s the primary experimental approach for research concerning the e f f e c t s of a i r p o l l u t i o n on p l a n t s . Models that r e l a t e O3 concentrations to crop y i e l d s are necessary for quantifying the e f f e c t s of O3 on crops growing at present ambient O3 concentrations and f o r the assessment of the e f f e c t s of present and a l t e r n a t i v e ambient a i r q u a l i t y standards for O3. A number of exposure-response models have been used by various authors u t i l i z i n g both l i n e a r and non-linear r e l a t i o n s h i p s . In the NCLAN program, the l i n e a r model was f i r s t selected over other models, as few of the r e l a t i o n s h i p s found were s i g n i f i c a n t l y improved by the use of higher order functions (Heck et a l . 1983). In an e f f o r t to combine data sets among c u l t i v a r s to produce a crop model for each crop species, the higher order models were reassessed and the Weibull model was ultimately chosen over l i n e a r , polynomial, plateau-linear and l o g i s t i c models (Heck et a l . , 1983). The reasons for i t s s e l e c t i o n include, 1) i t s o v e r a l l goodness of f i t and 2) i t s f l e x i b i l i t y i n covering the range of exposure-response r e l a t i o n s h i p s observed. Its proponents have also claimed the 20 i n t e r p r e t a t i o n of i t s parameters to be s t r a i g h t forward and b i o l o g i c a l l y meaningful (Heck et a l . , 1983; Rawlings and Cure, 1985). The basic Weibull y i e l d response model outlined by Rawlings and Cure (1985) i s : Y± = a texp[-( X i/CT ) c ] ] + e i (5) where Y^ i s y i e l d ; i s O3 exposure; the exponential portion of the model characterizes the proportional y i e l d response; "a" represents a hypothetical maximum y i e l d at zero O3 exposure (y i n t e r c e p t ) ; "cr " i s the exposure causing y i e l d to be reduced to 0.37a; "c" controls the shape of the response curve (large values of c give a plateau e f f e c t to the response curve); and ej^ i s random e r r o r . .The model has been extensively used by NCLAN for c a l c u l a t i n g predicted y i e l d s f o r a number of season-long 7-h or 12-h, d a i l y mean O3 concentrations (M7 and M12 r e s p e c t i v e l y ) . Nosal (1984), i n recognition of the importance of peaks and the number and duration of episodes, incorporated the three parameters ou t l i n e d i n Section 2.3 i n t o the following best- f i t t i n g p r o b a b i l i s t i c model; a mixed mu l t i v a r i a t e polynomial-Fourier regression model: R =f L + a i ( n ) x i n +LL + b i ( n 0 s i n v i <6> i=l n i=l m where R = plant response; X 1 = the number of pollutant episodes (days where the concentration exceeded 1 ppb); X 2 = the i n t e g r a l of concentration over each episode accumulated over the experiment (concentration and duration, pbb.h); X 3 = the s i n g l e peak po l l u t a n t concentration during the exposure period; and a ^ n V and b^ m^ are model parameters estimated using 21 the l e a s t squares approach. While the basis of t h i s model addresses the concern of both e p i s o d i c i t y and magnitude of exposures, the occurrence of s i g n i f i c a n t Fourier c o e f f i c i e n t s when applied to y i e l d data from soybeans exposed to 0 3 and/or S0 2 i s not e a s i l y accounted for i n b i o l o g i c a l terms (Nosal, 1984). The most serious of the l i m i t a t i o n s of the exposure-response models developed to date i s that many of them were developed from data based on steady-state a r t i f i c i a l p o l l u t i o n exposures. Work by Musselman et a l . (1983) and Hogsett et a l . (1985) has shown that the d i s t r i b u t i o n of po l l u t a n t concentrations during exposure a f f e c t s plant response. They found that episodic rather than constant or d a i l y peak exposure p r o f i l e s with the equivalent integrated exposure had a greater e f f e c t on plant response to 0 3. Musselman et a l . (1983) further suggested that studies u t i l i z i n g a constant (uniform) d i s t r i b u t i o n may be underestimating the magnitude of y i e l d responses to ambient p o l l u t a n t s . In t h e i r experiment, t o t a l duration of exposure was the same for both d i s t r i b u t i o n treatments. Musselman et a l . (1986) further examined the e f f e c t of peak concentration with two d i s s i m i l a r temporal d i s t r i b u t i o n s on plant response. In t h i s experiment with both t o t a l dose and peak concentration held constant by varying the duration of the exposures, no d i f f e r e n c e i n plant response to the two d i s s i m l a r d i s t r i b u t i o n s was found. However, the length of the fumigation period was longer for the simulated ambient d i s t r i b u t i o n with a c e n t r a l peak concentration than the exposure duration of the treatment i n which the concentration was held s t e a d i l y at the peak concentration (2.3 h). Lefohn and Runeckles (1987) point out that, i n terms of impact on the bean plants studied, t h i s r e a l l y amounts to a broad peak and hence i t i s not s u r p r i s i n g that comparable e f f e c t s were observed. 22 In contrast Heagle et a l . (1987) compared the response of Nicotiana  tanacum L. (tobacco) to both 7- and 12- h proportional versus constant additions of 0 3 to open-top chambers. Over the season, peaks occurred more frequently and were of greater magnitude for the proportional than the constant rates of addition of O3. However the seasonal means were s i m i l a r and there were no s i g n i f i c a n t differences i n y i e l d . Nevertheless y i e l d was 10% less i n the 12-h proportional than the 7-h proportional treatment. These r e s u l t s r e f l e c t those found by Heagle et a l . (1986) for two years of work with Glycine max L. cv. Davis (soybean). In t h i s experiment they compared response models using two exposure s t a t i s t i c s , integrated exposure (ppm.h) and the M7 seasonal mean, and found a d i s p a r i t y between predicted y i e l d reductions for the two years with the ppm.h s t a t i s t i c . However, t h e i r conclusion that the use of a season-long mean was j u s t i f i e d because i t gave consistent r e s u l t s for both years has been challenged (Runeckles, 1988). One would not expect the y i e l d reduction to be the same from season to season because d i f f e r e n t ambient conditions prevai l e d , i n addition to v a r i a t i o n s i n c u l t u r a l techniques, s o i l conditions, etc., and the lengths of the exposure periods used. This work i s i n contrast with that of Male et a l . , (1983) who compared the y i e l d of a number of crops exposed to both constant S0 2 concentration and a stochastic time s e r i e s . They found that a constant concentration treatment underestimated y i e l d losses compared with the stochastic time s e r i e s . 2.5 E f f e c t s of ozone on crops Various e f f e c t s of ozone on crops are well documented i n the l i t e r a t u r e . Most of the e a r l y work evaluated the e f f e c t of various concentrations of ozone on the production of f o l i a r symptoms of i n j u r y , 23 necrosis, etc. (Davis and Kress, 1974; Rajput and Ormrod, 1976; Reinert et a l . , 1984). The concentrations used i n these studies were r e l a t i v e l y high, ranging from 0.25 to 0.6ppm and above for short periods of time, which are more representative of acute as opposed to chronic exposure. Furthermore exposures were c a r r i e d out i n growth chambers and greenhouses, and few crops were grown to maturity. Several studies have explored the r e l a t i o n s h i p between f o l i a r i n j u r y and y i e l d . Work by Oshima et a l . (1977a) found no d i r e c t r e l a t i o n s h i p for Lycopersicon esculentum L. (tomato). Heagle et a l . (1979a) found that the threshold O3 concentration for f o l i a r i n j u r y (0.02 to 0.07 ppm) i n Zea mays L. ( f i e l d corn) hybrid Coker 16 was lower than that required to reduce kernel y i e l d s (0.11 to 0.15 ppm). F o l i a r i n j u r y may well be a d i r e c t i n d i c a t o r of y i e l d reduction with those crops whose marketable y i e l d i s the leaves. Heagle et a l . (1979d) found the threshold season -long 7-h mean for f o l i a r i n j u r y on Spinacia  oleracea L. (spinach) c u l t i v a r s was between 0.02-0.06 ppm whereas 0.06-0.10 ppm was needed for s i g n i f i c a n t reductions i n shoot growth. Temple et a l . (1985), exposed Hordeum vulgare L. (barley) cv. Poco and CM-72 to a range of ozone concentrations i n open-top chambers and found that neither c u l t i v a r showed s i g n i f i c a n t f o l i a r i n j u r y or y i e l d responses. They suggested that the threshold for e f f e c t s of ozone was above the seasonal 7-h mean of 0.06 ppm. Heagle et a l . (1979c), i n v e s t i g a t i n g the s e n s i t i v i t y of Triticum aestivum L. (winter wheat) c u l t i v a r s to ozone using open-top chambers, found that the threshold concentration for s i g n i f i c a n t i n j u r y and reduced growth and y i e l d was 0.06-0.10 ppm across a l l c u l t i v a r s . Plants subjected to seasonal means of 0.10 ppm and 0.13 ppm had 16% and 33% less seed weight r e s p e c t i v e l y than those exposed to 0.03 ppm. 24 Kress et a l . (1985) found that c u l t i v a r s of wheat d i f f e r e d i n s e n s i t i v i t y to O3. Kohut et a l . (1987) found a s i g n i f i c a n t reduction i n y i e l d of Triticum aestivum L. (winter wheat), and noted that the Weibull exposure-response curves for d i f f e r e n t years were q u a l i t a t i v e l y d i f f e r e n t , i n d i c a t i n g that environmental factors and the length of the exposure period a l t e r e d the response to O3. Kress and M i l l e r (1984) found s i g n i f i c a n t reductions i n y i e l d of Zea  mays L. ( f i e l d corn). The Weibull response model f i t t e d to the data predicted a y i e l d loss of 2.6% at 0.06 ppm and 16.3% at 0.10 ppm r e l a t i v e to 0.025 ppm 03 > S i g n i f i c a n t reductions i n y i e l d of sweet corn have also been found by Oshima (1973) and Heagle et a l . (1972). Temple et a l . (1986) found severe f o l i a r i n j u r y i n Lactuca sativa L. (head lettuce) cv. Empire with a season-long 7-h mean ozone concentration of 0.104 and 0.128 ppm. These l e v e l s a l s o decreased marketable s i z e head weight 21% and 80% compared to growth i n CF a i r . The threshold for a s i g n i f i c a n t y i e l d reduction i n head weight was greater than 0.083 ppm. Kress and M i l l e r (1983) found a l i n e a r reduction i n seed weight i n Glycine max (L.) Merr. cv. Corsoy with increasing ozone concentration. Amundson et a l . (1986) a l s o found a l i n e a r decrease i n soybean y i e l d with increasing ozone concentration. In addition, they found that ozone delayed onset i n flowering. Water-stressed plants had a smaller percent reduction i n y i e l d a t t r i b u t a b l e to ozone. They speculated that water stress caused the plants stomates to remain closed, thus reducing the e f f e c t i v e O3 dose received. In contrast, Smith et a l . (1987) using ethylenediurea (EDU), a chemical protectant against O3, found no s i g n i f i c a n t d i f f e r e n c e s i n y i e l d between EDU-treated and untreated soybeans i n extensive f i e l d plantings, 25 despite season-long 7-h/day mean O3 concentrations of 0.06 ppm and greater. Based upon NCLAN crop loss models, these concentrations should have reduced y i e l d by 10-20 per cent. The timing of exposure to a i r p o l l u t a n t s has a l s o been shown to be important i n determining a plant's s e n s i t i v i t y . In general, the younger the plant at the time of exposure the greater the s e n s i t i v i t y (Reinert and Henderson, 1980). Intermittent exposure of tomato seedlings to ozone concentrations of 0.10 ppm for 8 hours or 0.40 ppm for one or two hours p r i o r to being planted out i n the f i e l d r esulted i n a reduction i n t o t a l marketable and U.S. No. 1 y i e l d i n three of four c u l t i v a r s i n the 0.40 ppm for two hours treatment (Henderson and Reinert, 1979). Time of day during which the exposure takes place i s a l s o important i n determining a plant's response to ozone. Tomato plants exposed to O3 i n the afternoon as opposed to the morning developed more i n j u r y (Reinert et a l . , 1972a). Responses were c u l t i v a r - s p e c i f i c and newly expanded leaves were more s e n s i t i v e than young immature leaves. Menser et a l . (1963) also found that 0 3 - s e n s i t i v i t y i n Nicotiana glutinosa L. was i n v e r s e l y r e l a t e d to leaf maturity and that recently mature leaves were most s e n s i t i v e . This was also found by Ting and Dugger (1968) working on cotton leaves. Furthermore cotton leaves were most s e n s i t i v e to O3 before f u l l expansion during the period corresponding to a depletion of carbohydrates and amino acids i n the leaves (Ting and Mukerji, 1971). They suggested that repair of damage at t h i s stage cannot take place due to the depletion of soluble reserves. Ozone also a f f e c t s the p a r t i t i o n i n g of assimilates i n p l a n t s . Thus the treatment e f f e c t may not be expressed uniformly throughout the plant. 26 The root weight to t o t a l dry weight r a t i o and root to shoot r a t i o of carrots decreased s i g n i f i c a n t l y when exposed i n t e r m i t t e n t l y to 0.19 or 0.25 ppm 0 3 throughout t h e i r growth (Bennett and Oshima, 1976). Oshima et a l . (1978) found a decrease i n the root to shoot r a t i o f or Petroselinum crispurn ( m i l l . ) Nym. (parsley) cv. Banquet, following intermittent exposures to 0.20 ppm O3 throughout growth. Tingey et a l . (1971) found a greater reduction i n root fresh weight than leaf fresh weight for radishes and speculated that exposure to 0.05 ppm ozone may impair leaf metabolism r e s u l t i n g i n a reduction of photosynthate, or i n t e f e r e with t r a n s l o c a t i o n . Oshima et a l . (1979) observed that exposure of Gossypium hirsutum L. (cotton) cv. A l c a l a SJ-2 to an ozone concentration of 0.25 ppm reduced both vegetative growth and b o l l production. Furthermore the greatest reduction i n dry weights of a l l plant parts were found for the roots and b o l l s . Bennett and Runeckles (1977) a l s o found a reduction i n the root to shoot r a t i o for Lolium multiflorum Lam. (annual ryegrass) and T r i f o l i u m  incarnatum L. (crimson clover) exposed to O3 concentrations of 0.03 and 0.09 ppm. Blum et a l . (1983) exposed T r i f o l i u m repens L. (Ladino clover) cv. Tillman to a range of ozone concentrations (0 to 0.15 ppm) for four hours d a i l y for s i x days, and found that exposure to ozone resulted i n a greater reduction i n root dry weight (42%) than shoot dry weight (24%). The proportion of carbon a l l o c a t e d to the roots decreased 21% at 0.15 ppm and for the shoots, increased 79%. At concentrations l e s s than or equal to 0.10 ppm, carbon a l l o c a t e d to the roots increased from 35 to 52% and for the leaves decreased from 64 to 48%. The authors suggested that the more negative impact on the roots was a r e s u l t of a l t e r e d metabolism of the photosynthetic surfaces associated with O3 i n j u r y and reduction i n t r a n s l o c a t i o n of photosynthates to the roots. 27 Okano et a l . (1984) investigated the e f f e c t of continuous exposure to O3 at 0.2 ppm for four days, on assimilate p a r t i t i o n i n g i n Phaseolus  v u l g a r i s L. plants. They found the l a b e l l e d assimilates exported from the primary leaf (which acts as the main source of photosynthates for growth of the roots) decreased i n the presence of O3 due to both a reduction i n C0 2 f i x e d (62%) and i n h i b i t i o n of t r a n s l o c a t i o n . However, the amount of a s s i m i l a t e from the f i r s t t r i f o l i a t e leaf (which p r i m a r i l y supplies the immature growing leaves) only had a 24% decrease i n C0 2 f i x e d and t h i s was * compensated by an increase i n t r a n s l o c a t i o n . The o v e r a l l consequence was a greater proportion of a s s i m i l a t e being p a r t i t i o n e d to the growing leaves at the expense of the root and stem. In the presence of O3 the amount of l a b e l l e d a s s i m i l a t e translocated to the stem and roots decreased by 53% while that translocated to the leaves decreased by 28%. They suggested that the plants had adapted themselves to a p o l l u t e d environment such that e f f e c t s on growth were minimized. The change i n d i s t r i b u t i o n of dry matter during crop development, caused by O3 may a i d i n explaining the poor as s o c i a t i o n between f o l i a r symptoms and y i e l d reductions (Jacobson, 1982). 2.6 Environmental e f f e c t s S e n s i t i v i t y to ambient O3 concentrations has been r e l a t e d to a number of environmental f a c t o r s . Primary among these are water stress and r e l a t i v e humidity, both of which a f f e c t stomatal function (Rich and Turner, 1972; Kobriger and T i b b i t t s , 1985). Other environmental factors such as l i g h t i n t e n s i t y , temperature, a i r v e l o c i t y and s o i l f e r t i l i t y have also been shown to influence plant response to a i r pollutants (Heck et a l . , 1965). Ting and Dugger (1986) found that susceptible leaves were only s e n s i t i v e to ozone following a few hours i n the l i g h t and l o s t s e n s i t i v i t y 28 towards the end of the day. Dunning and Heck (1977) found that Pinto bean plants grown at high l i g h t i n t e n s i t i e s were less s e n s i t i v e to 0 3 across a l l exposure i n t e n s i t i e s than plants grown at lower l i g h t i n t e n s i t i e s and showed a s i g n i f i c a n t increase i n f o l i a r i n j u r y across a l l exposure l i g h t treatments. Kobriger and T i b b i t t s (1985) found that exposure to high r e l a t i v e humidity p r i o r to ozone exposure increased i n j u r y . This was associated with an increase i n stomatal conductance i n these p l a n t s . Dunning and Heck (1973) found a s i g n i f i c a n t increase i n ozone-induced f o l i a r i n j u r y i n pinto bean with increasing humidity during exposure to O3. Plants grown at high humidity (75-90%) were more s e n s i t i v e at a l l exposure humidity treatments than those grown at low humidity (45-60%). Otto and Daines (1969) also noted that O3 i n j u r y on Pinto beans increased with increasing humidity (21-95% at 31 °C). They found a c o r r e l a t i o n between the degree of i n j u r y and stomatal aperture, both of which increased with increasing humidity. McLaughlin and Taylor (1981) observed a nohproportional uptake of O3 with increasing O3 concentration at high humidity (75%) for Phaseolus  v u l g a r i s (Bush Blue Lake var. 274). F o l i a r uptake of O3 increased with an increase i n r e l a t i v e humidity from 35% to 75%. At low r e l a t i v e humidity, O3 uptake decreased as the O3 concentration increased. They found that changes i n i n t e r n a l leaf resistance were p r i m a r i l y responsible for reduced O3 uptake at the higher O3 concentrations as r e l a t i v e humidity had a greater e f f e c t on the conductance r a t i o s than on photosynthesis. Furthermore t o t a l leaf resistance to O3 increased 46% at 75% humidity and 83% at low humidity as the O3 was increased from 0.075 to 0.140 ppm. Therefore i t seems that a l t e r e d rates of pollutant uptake are the'main cause of humidity-induced v a r i a t i o n i n plant s e n s i t i v i t y to O3 and the 29 mechanism for these di f f e r e n c e s was a change i n i n t e r n a l leaf resistance as opposed to stomatal regula t i o n . They further suggested that these di f f e r e n c e s may be a function of the influence of humidity on water f l u x through the l e a f . Reich et a l . (1985) found a l i n e a r decrease i n leaf conductances with increased concentrations of 0 3 (0.01 to 0.13 ppm). Rich and Turner (1972) observed a rapid increase i n stomatal resistance i n water-stressed bean plants (Pinto). Well-watered plants responded more slowly i n the presence of O3. High humidity caused no change i n the stomatal resistance of the bean plants whereas stomatal resistance increased i n a dry atmosphere with exposure to 0 ^ The work of Tingey and Hogsett (1985) on Pinto bean suggested that water stress protects the plant from 0 3 i n j u r y p r i m a r i l y through i t s a c t i o n on the stomata rather than by biochemical or p h y s i o l o g i c a l changes i n the l e a f . However, Heggestad et a l . (1985) found a greater reduction i n y i e l d of soybean with water-stressed plants than those not subjected to water stress i n NF a i r plus O3 versus CF a i r . In contrast Temple et a l . (1985) found O3 had no e f f e c t on the y i e l d of water-stressed cotton i n 1981 experiments, but reduced the y i e l d of i r r i g a t e d cotton by up to 45% at twice the ambient O3 concentration (0.125 ppm) r e l a t i v e to CF a i r c o n t r o l s . However, i n 1982, the y i e l d of both water-stressed and non-stressed cotton plants to O3 was s i m i l a r , decreasing by 65% at twice ambient O3 concentrations (0.077 ppm) r e l a t i v e to c o n t r o l s . They speculated that the cooler, more humid growing conditions i n that year increased the s u s c e p t i b i l i t y of the plants to 0 3. Plants grown at high temperatures (27-32 °C) were found to be more s e n s i t i v e to O3 across a range of exposure temperatures (21-32 °C) than those grown at low temperature (21 °C) (Dunning and Heck, 1977). 30 Heagle et a l . , (1971) found that the e f f e c t of a i r v e l o c i t y during exposure to O3 appeared to be species s p e c i f i c . Bean was more severely injured at high (0.82 to 1.1 km/hr) than low (0.26 km/hr) a i r v e l o c i t y , but a i r v e l o c i t y had l i t t l e e f f e c t on oats or cucumber. Ashenden and Mansfield (1977) found that exposure of Ryegrass to S 0 2 at a windspeed of 25 m m i n - 1 s i g n i f i c a n t l y decreased leaf area, root to shoot r a t i o and a l l dry weights measured whereas at 10 m m i n - 1 no reductions i n growth were found. It i s f e l t that the s e n s i t i v i t y of a plant to a i r po l l u t a n t s at varying wind speeds i s r e l a t e d to di f f e r e n c e s i n boundary layer resistance of the leaves. Hence, a i r v e l o c i t y may be a c r i t i c a l f a ctor i n chamber studies, since, at low v e l o c i t i e s , the appreciable boundary layer resistance present requires a greater concentration of pollutant to achieve f l u x rates (and hence e f f e c t i v e doses) comparable to those r e s u l t i n g from lower concentrations i n situ a t i o n s i n which higher v e l o c i t i e s reduce the boundary layer r e s i s t a n c e . Heagle et a l . (1983a) compared the e f f e c t s of chronic doses of O3 on Glycine max (soybean) cv. Davis growing i n the ground versus pots containing a mixture of top s o i l , sand, p e r l i t e , peat moss, and v e r m i c u l i t e . They found no e f f e c t of growing media on the percent y i e l d loss due to O3 and suggested that plant response to O3 would be f a i r l y uniform over a wide range of substrate types. In contrast Heagle et a l . (1979c) found a greater reduction i n the y i e l d of Trit i c u m aestivum L. (winter wheat) exposed to O3 i n the ground versus those plants exposed i n pots. In each case open-top chambers were used to dispense the O3 treatments. A number of researchers have examined the i n t e r a c t i o n between O3 and f e r t i l i z e r on plants. However, the data i n the l i t e r a t u r e are not 31 consistent and the r e l a t i o n s h i p s between s o i l f e r t i l i t y and plant response to 0 3 remain unclear. Leone (1976) found that potassium d e f i c i e n c y i n tomato plants reduced O3 i n j u r y and that t h i s was r e l a t e d to an increase i n stomatal resistance i n the d e f i c i e n t p l a n t s . Leone and Brennan (1970) found an increase i n i n j u r y i n tomatoes with an increase i n the amount of phosphorus supplied. They suggested that the carbohydrate reserve accumulated i n phosphorus-deficient plants may p a r t l y explain the resistance of these plants to O3 exposure. 2.7 Exposure methodology The development of exposure-response r e l a t i o n s h i p s which are d i r e c t l y a p p l i c a b l e to crops growing under f i e l d conditions requires that plants be exposed to a range of pollutant exposure regimes under conditions that resemble those t y p i c a l of, or extrapolatable to, those occurring i n the f i e l d . Much of the early work on the e f f e c t s of a i r pollutants on plants was conducted i n growth chambers and greenhouses. Lewis and Brennan (1977), evaluated the response of Phaseolus v u l g a r i s cv. Pinto U.I. 114 following exposure to elevated ozone concentrations i n the greenhouse, growth chambers and open-top f i e l d chambers. Based on symptomatology and percent f o l i a r i n j u r y they concluded that s i g n i f i c a n t d i f f e r e n c e s occur i n the ozone response of greenhouse and growth chamber plants compared to the open-top chamber plants. In a l l tests the greenhouse plants were most s e n s i t i v e and plants i n open-top chambers the l e a s t . In ad d i t i o n , symptoms that developed i n the open-top chamber plants were representative of those commonly observed on f i e l d grown plants, whereas the symptomatology of greenhouse and growth chamber plants was frequently a t y p i c a l . 32 Hucl and Beversdorf (1982) screened a number of Phaseolus v u l g a r i s L. c u l t i v a r s f or s e n s i t i v i t y to ozone, using both growth chamber and f i e l d p l o t s . They found that a l l bean c u l t i v a r s were s e n s i t i v e to ozone under growth chamber conditions. However, under f i e l d conditions, late-maturing v a r i e t i e s were not injured by ambient ozone l e v e l s . De Vos et a l . (1983) i n a comparison of potato genotypes grown i n growth chambers versus the f i e l d found that those c u l t i v a r s d i s p l a y i n g s u s c e p t i b i l i t y to oxidants i n the f i e l d behaved s i m i l a r l y i n the laboratory; however c u l t i v a r s that were susceptible i n the laboratory frequently were r e s i s t a n t i n the f i e l d . In contrast, Meiners and Heggestad (1979) found a s i g n i f i c a n t c o r r e l a t i o n (r = 0.20) between ozone-induced f o l i a r i n j u r y on mature f i e l d grown plants of Phaseolus v u l g a r i s L. (snap bean) and seedlings exposed to ozone i n growth chambers. The e f f e c t s were c u l t i v a r - s p e c i f i c . However, i n view of the o v e r a l l evidence a v a i l a b l e , meaningful extrapolation from laboratory studies to f i e l d grown crops i s far from straight forward. There are a number of approaches i n use for assessing the impact of a i r p o l l u t i o n on crops i n f i e l d s i t u a t i o n s . These include open-top chambers (Mandl et a l . , 1973; Heagle et a l . , 1973), zonal a i r p o l l u t i o n systems (ZAPS) (Lee and Lewis, 1977; Runeckles et a l . , 1981a), the use of open a i r , natural pollutant gradients (Oshima et a l . , 1976, 1977b) and the use of ethylenediurea (EDU) as an antioxidant i n the f i e l d (Carnahan et a l . , 1978; Adomait et a l . , 1987; Smith et a l . , 1987) Since growth chambers have been shown to be inadequate for assessing the e f f e c t s of 0 3 on crops under f i e l d conditions, chambers of various types were developed for exposing plants i n the f i e l d . These chambers were designed to minimize interference with plant growth, simulate f i e l d conditions and allow control of the composition of the a i r surrounding the 33 p l a n t s . NCLAN and other studies have made extensive use of open-top chambers for t h i s purpose (Heck et a l , 1983). However, open-top chambers have been shown to modify the plant environment i n various ways. For example, a i r temperatures are higher, l i g h t i n t e n s i t y i s lower (Heagle et a l . , 1979a) and evaporative water loss i s less i n the chambers (Olszyk et a l . , 1980). Olszyk et a l . (1980), Heagle et a l . (1979a), and others have a l s o found that plants i n open-top chambers grow d i f f e r e n t l y i n d i f f e r e n t parts of the chamber. Their most serious drawback thus concerns various modifications of the microclimate around the plants which a f f e c t growth and whose e f f e c t s are confounded with the e f f e c t s of the po l l u t a n t treatments applied. A f i e l d exposure system which avoids these problems i s the Zonal A i r P o l l u t i o n System (ZAPS) described by Lee et a l . (1975) and, i n a modified version, by Runeckles et a l . (1981a). This approach involves the release of pollutant from o r i f i c e s i n a network of pipes supported above the ground, and r e l i e s on d i f f u s i o n and turbulance to mix the released gas with ambient air. to achieve a range of concentrations. Modification of the composition of the ambient a i r surrounding the crop i s achieved without the microclimatic e f f e c t s imposed by chambers. A number of drawbacks with t h i s type of system were pointed out by Krupa, (1984). These include: small changes i n windspeed w i l l produce extreme changes i n the O3 concentration; the need for p r e c i s i o n i n the system c o n t r o l l i n g the release of O3; the need for intensive monitoring of the O3 concentrations within the study p l o t ; and r e s t r i c t i o n of the a p p l i c a b i l i t y of the method to areas i n which the natural ambient 0 3 l e v e l s are low. An a l t e r n a t i v e open-air, chamberless approach u t i l i z e s a natural environmental gradient of ambient O3 concentrations. This approach has 34 been used i n Southern C a l i f o r n i a (Oshima et a l . , 1976, 1977b). Crops were grown i n f i e l d p l o t s with standard s o i l and water and maintained at d i f f e r e n t distances from a strong urban oxidant source. The drawbacks of t h i s approach are similar to those of the ZAPS approach i n that extensive and i n t e n s i v e monitoring are necessary to accurately characterize the exposures to which the plants are subjected. However, i t s value l i e s i n the use of r e a l world p o l l u t a n t f l u c t u a t i o n s . The use of EDU i n assessing crop y i e l d losses due to 0 3 has been shown by a number of researchers to be a r e l i a b l e a l t e r n a t i v e to the other f i e l d techniques (Bisessar, 1982; Clarke et a l . , 1983; Foster et a l . , 1983a; Smith et al.,1987). However concerns have been ra i s e d regarding the e f f e c t of EDU on vegetation and the degree of protection from oxidants that i t provides (Lefohn and Runeckles, 1987). 35 3. MATERIALS AND METHODS-1985 EXPERIMENT 3.1 EXPERIMENTAL DESIGN The experiment was conducted using the Zonal A i r P o l l u t i o n System (ZAPS) located on the Totem F i e l d of the Department of Plant Science, U n i v e r s i t y of B r i t i s h Columbia. The treatment blocks were 10 by 12 m located 25 m apart to minimize possible carry-over e f f e c t s (Figure 2). Ozone treatments were assigned to the treatment p l o t s marked AAx2.5, AAx2.0 and AAxl.5, on the basis of the p r e v a i l i n g wind. The fourth block (AA) served as a c o n t r o l . The c e n t r a l 8 x 10 m area of each block, within a i m border, was subdivided i n t o 1 by 2 m p l o t s and 2 c u l t i v a r s each of lettuce, pea, bean, carrot and b r o c c o l i were randomly assigned to 4 p l o t s per treatment. The border surrounding each block was sown with the crop i n the adjacent p l o t . Plants were not harvested from border p l o t s . In t h i s experimental design, with each O3 treatment a l l o c a t e d to a s p e c i f i c block, no true r e p l i c a t i o n of the treatments was p o s s i b l e . Although the p l o t s of the d i f f e r e n t crops were randomly d i s t r i b u t e d throughout each block, t h e i r use d i d not overcome any p o s s i b l e block X treatment i n t e r a c t i v e e f f e c t s . However, experience gained with the system, using SO2 on a v a r i e t y of species (Runeckles et a l . , 1981b), had shown that such i n t e r a c t i o n s were minimal. 3.2 FIELD PLOTS AND GAS DELIVERY SYSTEM The ZAPS gas d e l i v e r y system previously described by Runeckles et a l . (1981a) consisted of a manifold of PVC pipes supported by aluminum posts 80 cm above the ground over each of the three treatment blocks (Figure 2). No manifold was used over the fourth block, which served as a c o n t r o l . The release o r i f i c e s , h o r i z o n t a l l y positioned on a l t e r n a t i n g sides of the Figure 2: Layout of treatment blocks and network of a d i s t r i b u t i o n manifold of the ZAPS at U.B.C. AA = ambient a i r ; AAxl.5 = ambient a i r x 1.5; AAx2.0 = ambient a i r x 2.0 and AAx2.5 = ambient a i r x 2.5. 37 / 25 m 10 SUPPLY AND INSTRUMENTATION PREVAILING WINDS l ! AAx1.5 I 25 m FENCE LINE 3m f - "1 SUPPLY LINE 3m J L * * t < 1 L « f • < ' i i | 1 « V 1 i < 1 • • i 4 • \ * • * 1 i — i i • « > i » 1 1 ORIFICES SAMPLING LOCATION 38 pipes, 1.0 metre apart, were 0.8 mm i n diameter. The three d e l i v e r y manifolds had the same number of release o r i f i c e s . These manifolds were continuously supplied with a i r (which could be enriched with 0 3 ) , through underground PVC pipes, 24 hours per day for the season, from a Becker SV2.330/2 Vortex periphery blower. The ozone-generating system consisted of a Grace Model GS 4060 ozone generator supplied with compressed oxygen. Ozone was supplied over a 14-hour period, 0700-2100 h ( P a c i f i c daylight time, (PDT)) commencing July 23 and ending September 27, 1985. 3.3 OZONE MONITORING AND CONTROL The four treatments were ambient a i r (AA) and three treatments i n which ozone was added i n proportion to the ambient ozone concentration: AAxl.5; AAx2.0 and AAx2.5. The gas release system thus simulated the d i u r n a l cycle t y p i c a l of ozone by tracking the ambient ozone concentration. Monitoring information derived from the control p l o t was used to regulate the output of the ozone generator. This control was achieved through a program run by the Campbell 21-X data logger used f or data a c q u i s i t i o n , which regulated the voltage output to a servomotor which, i n turn, adjusted the voltage of the ozone generator, and thus i t s output. The d i f f e r e n t l e v e l s of ozone enrichment on the d i f f e r e n t blocks were achieved by unequal s p l i t t i n g of the flow of ozone released into the pipes supplying the manifolds. D i l u t i o n and mixing of the 0 3 released occurred as a r e s u l t of d i f f u s i o n and turbulence. An ozone monitor (Model 1003-AH, Dasibi Environmental Corporation Glendale, CA.) operated on a time-sharing basis among 8 monitoring points (2 per block). Teflon a i r sample l i n e s (1/4 inch OD) of equal length from 39 each block were enclosed i n PVC p i p i n g and run to the monitoring hut underground (Figure 2). A i r was drawn continuously through the l i n e s from each monitored point (40 cm above ground l e v e l ) and sequentially diverted to the monitor. The diverted a i r stream passed through the ozone monitor for four minutes every ten minutes for the AAx2.5 treatment and for two minutes every ten minutes for the AA, AAxl.5 and AAx2.0 treatments. Hence, each sample l o c a t i o n f or a l l 4 blocks was sampled once every 20 minutes. Feedback from the AA and AAx2.5 sample l i n e s was used to adjust the output to the ozone generator every 10 minutes. While the AA sample readings c o n t r o l l e d the ozone generator output, i f ozone concentrations exceeded 250 ppb i n the AAx2.5 treatment, feedback from the Campbell 21-X data logger turned the generator off u n t i l the next reading below the l i m i t was recorded. This over-ride was provided to prevent excessive O3 build-up i n s t i l l - a i r conditions. As the system was driven by ambient ozone concentrations, exposures were minimal on overcast or rainy days i n a l l treatments. In order to e s t a b l i s h the nature of the v e r t i c a l d i s t r i b u t i o n of ozone, v e r t i c a l p r o f i l e s of ozone concentration were obtained for two sampling locations i n one d i s t r i b u t i o n manifold (AAx2.5 p l o t ) . This was done by measuring ozone concentrations for a 15 minute period at 10 cm v e r t i c a l i n t e r v a l s from 10 to 100 cm (45 i n d i v i d u a l O3 readings per l o c a t i o n were recorded on a chart recorder) over bare s o i l . 3.4 DATA HANDLING A l l data were recorded on a Campbell S c i e n t i f i c Model 21-X data a c q u i s i t i o n system for i n i t i a l processing and were stored on cassette tape, 40 p r i o r to transfer v i a a C20 cassette i n t e r f a c e to the U.B.C. mainframe computer for a n a l y s i s . 3.5 FIELD PREPARATION The land was r o t o t i l l e d followed by harrowing. Pre-season s o i l sampling indicated that the p l o t s were r e l a t i v e l y uniform for organic matter, s o i l pH, N, P and K. A l l p l o t s were limed with dolomite (65 mesh) at the rate of 4 tonnes/ha on June 18. An a l l purpose f e r t i l i z e r (13:16:10) was broadcast at the rate of 900 kg/ha, p r i o r to seeding. On June 24-26 the plo t s were raked by hand and the ZAPS was set up over the three treatment p l o t s . 3.6 PLANT MATERIALS The plant species used i n t h i s experiment were two c u l t i v a r s f o r each of Latuca s a t i v a L. (head le t t u c e , cv's. Ithaca and Montello), Pisum  sativum L. (pea, cv's. Puget and Bolero) Phaseolus v u l g a r i s L. (bean, cv's. BBL-GV2 and Galamore), Daucus carota L. (carrots, cv's. Hipak and Sixpak) and Brassica oleracea L. ( b r o c c o l i , I t a l i c a group cv's. Emperor and SGI). Pea and bean seeds were supplied by W. Brotherton Seed Co., Inc. Moses Lake, Washington, USA through Royal C i t y Foods Ltd.. A l l other seeds were supplied by the B.C. M i n i s t r y of A g r i c u l t u r e and Food (now B.C. M i n i s t r y of A g r i c u l t u r e and F i s h e r i e s ) . The crops and c u l t i v a r s were chosen based on vegetable production data from the l a s t 5 years which demonstrated t h e i r importance i n terms of d o l l a r value and the amount of production on the Lower Mainland (Data provided by M. Sweeney, B.C. M i n i s t r y of A g r i c u l t u r e and Food, Abbotsford). 41 The seeding (transplanting for b r o c c o l i ) rates for the crops were i n accordance with recommendations outlined i n the 1985 Vegetable Production Guide for B.C.. A l l seeding or planting took place on July 2-3 1985. Lettuce plants were thinned to 30 cm apart when true leaves had formed. 3.7 CROP AND PEST MANAGEMENT The p l o t s were examined d a i l y f o r pests, diseases, ozone i n j u r y symptoms and plant development during the experimental period. Lettuce, peas and beans had emerged by Ju l y 7 and carrots by Ju l y 14. Tensiometers ( S o i l Moisture Equipment Corp. Santa Barbara, CA.) were i n s t a l l e d J u l y 25 at 15 and 25 cm depths i n each block. I r r i g a t i o n water was applied July 13, 15, 18,31, and August 6, 12, 18, 19 and 27 by overhead sp r i n k l e r to maintain s o i l moisture at f i e l d capacity. A l l blocks received the same amount of i r r i g a t i o n water throughout the season. Aphids, cutworms, root maggots and cabbage moths were observed on a l l blocks. To control these pests, Diazinon was applied on the f o l i a g e when necessary at the recommended rates. 3.8 HARVEST SCHEDULE AND PLANT MEASUREMENTS The date of the harvest depended on the species and the time required for maturation (Table 1). At harvest, plants were randomly selected from the treatment p l o t s and a l l crops excluding carrots were cut at ground l e v e l . The number of plants taken v a r i e d with the crop: f o r peas and beans, eight plants; for l e t t u c e , three plants;, and for b r o c c o l i , f i v e p l a n t s . The number of carrot plants harvested varied with each p l o t as germination was e r r a t i c . Measurements taken included plant height, leaf 42 number, f l o w e r / f r u i t number and dry weights of a l l component par t s . A fresh weight-dry weight conversion factor was calculated f o r each crop to estimate marketable weight f or each type of produce (Appendix 1.). Leaf areas were obtained using a LI-COR LI-3100 leaf area meter. Dry weights of the component parts were obtained using a Mettler PC4400 balance a f t e r samples were d r i e d to constant weight i n a forced a i r oven at 70 °C. Table 1. L i s t i n g of crops and f i n a l harvest dates for the 1985 f i e l d experiment. Crop Harvest Date Growing season length (days) Exposure season length (days) B r o c c o l i August 25, 26 54 33 Bean September 12 72 51 Pea September 13 73 52 Carrot September 20,21 80 59 Lettuce September 28 88 67 3.9 DATA ANALYSIS Y i e l d data were analyzed by an a l y s i s of variance using a s p l i t p l o t design. The treatment sum of squares was p a r t i t i o n e d i n t o either l i n e a r or quadratic terms, but not both, due to the lack of a v a i l a b l e degrees of freedom. 43 To e s t a b l i s h the nature of the d i s t r i b u t i o n of ozone concentrations achieved for each treatment over the season, the hourly averages between 0700-2059 were grouped into concentration classes and p l o t t e d . Both normal and lognormal d i s t r i b u t i o n s were investigated. Exposure s t a t i s t i c s c u r r e n t l y employed i n the l i t e r a t u r e were computed from the concentration data for the development of the exposure response models. These included the season-long average concentration between 0900 and 1559 hours (M7); 0900 and 2059 hours (M12) and 0700 to 2059 hours (M14). In addition to these s t a t i s t i c s , the highest 1-hour peak during the season for the respective d a i l y time-frames was determined ( P l [ 7 ] , Pl[12] and Pl[14]); the maximum d a i l y 7-, 12- or 14-hour means for the season (P7, P12, and P14) and the average of the d a i l y maximum 1-hour means for the season (Ml[7], Ml[12] and Ml[14]). A l l of the above s t a t i s t i c s were determined by taking the averages of the hourly means for the respective time-frame over the growing season of each crop (Table 1). The complete l i s t i n g of the exposure s t a t i s t i c s computed i s presented i n Table 2. Simple l i n e a r and quadratic regression analyses were used to define the f u n c t i o n a l r e l a t i o n s h i p s between the marketable y i e l d of each c u l t i v a r and the exposure s t a t i s t i c s derived. Higher order models were not assessed due to the l i m i t e d number of points a v a i l a b l e i n the data sets. Regressions were calculated from the treatment means. Table 2. Exposure s t a t i s t i c s used as independent v a r i a b l e s i n l i n e a r and polynomial exposure-response models. Exposure Description Reference s t a t i s t i c M7 Season-long mean of d a i l y 7-h mean (0900-1559) M12 Season-long mean of d a i l y 12-h mean (0900-2059) M14 Season-long mean of d a i l y 14-h mean (0700-2059) Ml [7] Season-long mean of d a i l y 1-h max. (0900-1559) Ml [12] Season-long mean of d a i l y 1-h max. (0900-2059) Ml [14] Season-long mean of d a i l y 1-h max. computed between 0700-2059 h P7 Peak 7-h (0900-1559) season-long mean P12 Peak 12-h (0900-2059) season-long mean P14 Peak 14-h (0700-2059) season-long mean P l [7] Single peak 1-h mean (0900-1559) for the season P l [12] Single peak 1-h mean (0900-2059) for the season P l [14] Single peak 1-h mean (0700-2059) for the season Heck et a l . (1984) Heagle et a l , (1987) Heck et a l . (1984) Heck et a l , (1984) Heck et a l . (1984) 45 4. RESULTS-1985 EXPERIMENT 4.1 AIR QUALITY 4.1.1 S p a t i a l d i s t r i b u t i o n A v e r t i c a l ozone p r o f i l e generated over the AAx2.5 p l o t i n the absence of a plant canopy showed d i s p e r s a l of ozone down to the lowest point measured, 10 cm (Figure 3), thus confirming that the system was functioning as proposed, and providing the desired uniformity of concentrations. However, there was a c e r t a i n amount of v a r i a t i o n i n the average concentrations at a l l heights. 4.1.2 Temporal d i s t r i b u t i o n The frequencies of occurrence of d i f f e r e n t hourly O3 concentrations i n the 14-h period, 0700-2059, for ambient a i r and a l l treatments are shown i n Figure 4. Visual inspection of these histograms c l e a r l y shows that neither the ambient nor treatment p l o t s received normally d i s t r i b u t e d concentrations. A l o g - p r o b a b i l i t y p l o t of the cumulative ozone concentrations (ppb), for a l l treatments between 0700 and 2059-h i s presented i n Figure 5. Cumulative p r o b a b i l i t y r e f e r s to the p r o b a b i l i t y that an observtion w i l l be less than or equal to a p a r t i c u a l r value of Y. Since log-normally d i s t r i b u t e d data should r e s u l t i n a l i n e a r p l o t , the data presented c l e a r l y show that the d i s t r i b u t i o n s of concentrations are skewed and only poorly approximate log-normal d i s t r i b u t i o n s . Rather, they appear p l a t y k u r t i c , suggesting an inordinate number of ozone concentrations i n the mid-range. The exposure s t a t i s t i c used most widely i n the l i t e r a t u r e i s the season-long d a i l y 7-h mean (M7). This s t a t i s t i c was o r i g i n a l l y introduced by NCLAN to encompass that period of the day at which the 46 Figure 3: V e r t i c a l p r o f i l e s of mean ozone concentrations averaged over 2 locations (+ standard deviation), i n the AAx2.5 block. —> = manifold height (cm.) 47 E o, CJ ZD CO o CO UJ > o CD < I— zn 120 -i 1 0 0 -80 6 0 -4 0 -2 0 -[> D -L> -• -• -• •a •a •a -• 20 3 0 4 0 5 0 6 0 OZONE CONCENTRATION [ppb] I 70 48 Figure 4: D i s t r i b u t i o n of hourly ozone concentrations (ppb) for 0700-2059 hours for AA, AAxl.5, AAx2.0 and AAx2.5 treatments. 49 AMBIENT AIR AMBIENT AIR X 1.5 1800 1700-1600-1500-m 1 4 0 0 -z O 1300-5* 1200-£ "00-1 to m IOOO-o 9 0 0 " 800-700-600-500-400-300-200-100-0 >-a z LU 3 s rr i i i i i i 5 105 205 305 405 505 605 705 OZONE CONCENTRATION [ppb] AMBIENT AIR X 2.0 tn z o I or U i to GO o >-o Ld i i i I I I 5 105 205 305 405 505 605 705 OZONE CONCENTRATION [ppb] AMBIENT AIR X 2.5 z o § Ol UJ 00 m o u. O >-o 3 O 1800- 1800-1700- 1700-1600- 1600-1500- 1500-1400- to 1400-z 1300- o 1300-1200- 1200-1100- rr U l 1100-1000- to m 1000-O 900- n u. 900-800- 1 O 800->-700- UL a 700-z 600- U l 3 600-500- M o U l 500-400- ntnu or u. 400-300- H 300-200- Hi 200-100-0-100-0-5 105 205 305 405 505 605 705 OZONE CONCENTRATION [ppb] -i 1 1 r 5 105 205 305 405 505 605 705 OZONE CONCENTRATION [ppb] 50 Figure 5: Cumulative d i s t r i b u t i o n s of hourly ozone concentrations (ppb) for 0700 to 2059 hours for AA, AAxl.5, AAx2.0 and AAx2.5 treatments, p l o t t e d on l o g - p r o b a b i l i t y paper. 51 52 crops were considered to be most s e n s i t i v e , and the ambient a i r 0 3 l e v e l s highest (Heagle et a l . , 1987). The ranges of M7 values found i n the present study during the exposure period f or each of the species, across a l l treatments were: AA= 25.45-26.47 ppb; AAxl.5= 31.92-34.22 ppb; AAx2.0= 40.58-43.94 ppb; AAx2.5= 44.07-48.70 ppb (Appendix 2). Peak 1-h ( P l [ 7 ] ) , peak 7-h (P7) and mean d a i l y maximum 1-h ozone concentrations (Ml[7]), and comparable exposure s t a t i s t i c s f o r the 12-h and 14-h periods are also presented i n Appendix 2. The season-long pattern of average ozone concentrations achieved throughout the day i s shown i n Figure 6. Though the system was programmed to provide proportional additions, i t i s evident from these curves that t h i s was not f u l l y achieved. The i n a b i l i t y to track the early morning r i s e i n ambient O3 concentrations appears to have been the r e s u l t of a combination of overshoot and lag time i n the response, s t a r t i n g from r e s t . The i n a b i l i t y of the system to track the ambient 0 3 between 1700-2000 h was found to be due to a mechanical f a u l t i n the mechanism for adjusting the potentiometer c o n t r o l l i n g the voltage on the O3 generator. The servomotor was not powerful enough to c o n s i s t e n t l y turn the voltage s e t t i n g down when required, at t h i s point i n the day. 4.2 VISUAL OBSERVATIONS Neither head le t t u c e c u l t i v a r established well i n any of the treatment p l o t s ; hence the d e c i s i o n was made to disc a r d them from the experiment. By August 19, symptoms of acute and chronic ozone i n j u r y , manifested by severe c e l l collapse and s l i g h t bronzing re s p e c t i v e l y , were noted i n the AAx2.5 treatment for b r o c c o l i and bean leaves. Necrotic spots were also 53 Figure 6: Hourly seasonal means for AA, AAxl.5, AAx2.0 and AAx2.5 treatments. 55 observed on b r o c c o l i leaves i n the AAx2.0 treatment. These became more prevalant as the plants matured. Many of the leaves on the bean plants became c h l o r o t i c i n both the AAx2.0 and AAxl.5 treatments by August 2 and t h i s condition worsened as the plants matured. Examination of the bean roots at harvest indicated a root rot problem which was not consistent throughout a treatment or between treatments. B r o c c o l i roots were also found to be heavily i n f e s t e d with club root (Plasmodiophora brassicae Wor.) i n both the AAx2.0 and AAx2.5 treatments, with sporadic occurrences i n the AAxl.5 but none i n the AA treatment. Extensive powdery mildew (Erysiphe polygoni DC.) i n f e s t a t i o n s were evident on the AA treatment pea plants by August 28 and to a lesser extent i n the AAxl.5, AAx2.0 and AAx2.5 treatments. This condition worsened i n the AA treatment by September 6 and continued u n t i l the plants matured (September 13). Growth of the carrots i n the AAx2.0 treatment was better than expected. It became clear during the experiment that t h i s treatment p l o t d i d not dr a i n at the same rate as the other p l o t s . It i s suspected that the drainage t i l e s under t h i s p l o t were no longer functioning adequately. 4.3 ANALYSIS OF VARIANCE Results from the analysis of variance for b r o c c o l i (Table 3) demonstrate no s i g n i f i c a n t treatment e f f e c t s (p = 0.05), although examination of the data suggested there was a trend for a decrease i n a l l v a r i a b l e s measured, as demonstrated f o r the y i e l d v a r i a b l e s i n the regression analyses which follow i n Section 4.4. In contrast, dead leaf weight and number showed a reverse trend with increasing O3 exposure. S i g n i f i c a n t c u l t i v a r differences were found for height, l e a f number, dead 56 Table 3. Summary of anova results: F-values for the effects of ozone on broccoli variables (per plant) SOURCE df HT LN DLN LA FN OF VARIATION °3 0 3 LIN O3 QUAD O3 DEV CULTIVAR O3XCULT O3 LINXCULT O3 QUADXCULT O3 DEVXCULT (3) 1 1 3.65- S'. 97- 2.14- 2.19- 0.60-X 2 25.71* 2.47- 17.94- 4.64- 5.51-1 155.50* 23.12* 23.12* 1.15- 1.95-(3) 4.95- 2.49- 1.80- 1.52- 1.66-1 12.85- 5.47- 2.57- 2.97-1 3.42-2 0.OS- 1.25- 1.74- 0.51- 0.78-SOURCE df SW LW DLW FW MFW TW OF VARIATION °3 (3) O3 LIN 1 11.60- 10.01-O3 QUAD 1 O3 DEV 2 1.84- 3.85-CULTIVAR 1 2.87- 3.86-O3XCULT (3) 1.35- 3.42-O3 LINX CULT 1 8.03-O3 QUADXCULT 1 2.06-0 3 DEVXCULT 2 1.03- 0.56-2.30- 7.94- 12.91- 9.75-16.87- 8.69- 10.31- 4.70-0.16- 103.34** 220.22** 1.88-1.22- 1.44- 2.38- 1.38-2.31- 5.13- 2.15-1.65-0.66- 0.16- 0.09- 0.44-HT = Height (cm); LN = Leaf number; DLN = Dead leaf number; LA = Leaf area, (sq. cm), includes petiole; FN = Inflorescence number; SW = Stem weight (g); LW = Leaf weight (g); DLW = Dead leaf weight (g); FW = Inflorescence dry weight (g); MFW = Marketable inflorescence dry weight (g); TW = Total weight (g). = ns p = 0.05 ** p = 0.01 *** p = 0.001 57 leaf number, flower weight and marketable flower weight (Table 3). In t h i s experiment cv. SGI out performed cv. Emperor for a l l v a r i a b l e s measured, though not a l l d i f f e r e n c e s were s i g n i f i c a n t . Results from the a nalysis of variance for bean show a s i g n i f i c a n t quadratic treatment e f f e c t for marketable pod number, stem weight and t o t a l weight (Table 4); no other r e s u l t s for bean were s i g n i f i c a n t . S i g n i f i c a n t quadratic treatment e f f e c t s were also found for dead l e a f l e t number, t o t a l weight, and l e a f l e t weight and number i n pea, i n a d d i t i o n to c u l t i v a r d ifferences for these variables and pod number. There were a l s o s i g n i f i c a n t ozone x c u l t i v a r i n t e r a c t i o n s for pod weight, l e a f l e t , bud and pod number (Table 5). No s i g n i f i c a n t treatment e f f e c t s or c u l t i v a r d i f f e r e n c e s were found f o r carrot (Table 6). 4.4 REGRESSION ANALYSES 4.4.1 Simple Linear Regression Analyses The r e s u l t s of these analyses are shown i n Tables 7 through 10. No s i g n i f i c a n t l i n e a r regression models could be computed for either c u l t i v a r of pea and carrot or for Galamore bean. In no case d i d the use of the seasonal means (M7, M12 and M14) r e s u l t i n the best f i t t i n g l i n e a r response function. The best f i t t i n g models of y i e l d response to ozone concentration f o r those c u l t i v a r s y i e l d i n g s i g n i f i c a n t models are depicted i n Figures 7 through 9. For those c u l t i v a r s y i e l d i n g s i g n i f i c a n t regressions, the peak 1-h exposure s t a t i s t i c s resulted i n the best f i t s , as seen by the c o e f f i c i e n t s of determination and r e s i d u a l mean square (RMS). However, regardless of s i g n i f i c a n c e , i n a l l cases there was an inverse r e l a t i o n s h i p between ozone and y i e l d . Table 4. Summary of anova r e s u l t s : F-values for the e f f e c t s of ozone on bean va r i a b l e s (per plant) SOURCE OF VARIATION df HT LN DLN PN LA MPN °3 (3) O3 LIN 1 3.56-O3 QUAD 1 14.11- 13.47- 16.60- 14.70- 451.71** O3 DEV 2 0.93- 1.00- 1.50- 2.73- 0.92- 0.03-CULTIVAR 1 11.05- 0.27- 1.17- 2.99- 0.00- 4.26-O3XCULT (3) 2.32- 1.02- 2.83- 1.19- 0.76- 1.08-O3 LINX CULT 1 4.96- 1.24-0 3 QUADXCULT 1 1.05- 6.33- 1.57- 0.29-O3 DEVXCULT 2 0.74- 0.77- 0.46- 0.24- 0.65- 0.61-SOURCE df SW LW PW MPW DLW TW OF VARIATION 0 3 (3) O3 LIN 1 2 .5-O3 QUAD 1 17 .64* 14. 69- 3.24- 2.31- 19 .36* O3 DEV 2 0 .94- 0. 88- 2.05- 2.76- 2 .0- 0 .60-CULTIVAR 1 1 .40- 0. 17- 5.31- 5.99- 0 .5- 0 .06-O3XCULT (3) 0 .96- 0. 92- 0.78- 0.85- 1 .0- 0 .90-O3 LINXCULT 1 0 .89- 0. 76- 0.56- 0 .60-O3 QUADXCULT 1 0.34- 2 .0-O3 DEVXCULT 2 0 .74- 0. 62- 0.97- 0.99- 1 .0- 0 .79-HT = Height (cm); LN = Leaf number; DLN = Dead leaf number; PN = Pod number; LA = Leaf area (sq. cm , includes p e t i o l e ) ; MPN = Marketable pod number; SW = Stem weight (g) LW = Leaf weight (g); PW = Pod weight (g); MPW = Marketable pod weight (g); DLW = Dead leaf weight (g); TW = Tot a l plant weight (g) = ns * p = 0.05 ** p = 0.01 *** p = 0.001 59 Table 5. Summary of anova r e s u l t s : F-values for the e f f e c t s of ozone on pea var i a b l e s (per plant) SOURCE OF VARIATION df BW FW PW MPW DLW TW SW LW °3 (3) 0 3 LIN 1 3.3-0 3 QUAD 1 0.5- 2.3- 1.6- 0.2- 22 .0* 3.8- 57.8* O3 DEV 2 5 . 0 - 11 . 0 - 13 . 0 - 3.9- 4.5- 2.4- 1.7- 0.4-CULT 1 1 . 0 - 3.1- 4 . 0 - 0 .0 - 0.4- 24.1* 6.1- 16.5-O3XCULT (3) 1.8- 2.3- 8.6- 1.4- 1.3- 16.8- 3.2- 5.8-O3 LINXCULT 1 3.5- 1.1- 1.7- 15.4-O3 QUADXCULT 1 5 . 0 - 23.7* 48.4* 7.6-O3 DEVXCULT 2 0.5- 0.3- 0.3- 1.4- 0.3- 0 . 1 - 0.4- 0.26-SOURCE OF HT LN DLN BN FN PN LA MPN VARIATION °3 O3 LIN 0.7- 5.2-O3 QUAD 0.4- 1155.6*** 0.9- 0.6- 1.4- 6.1-O3 DEV 1.1- 0.0- 51.2* 3.5- 7.6- 4.2- 0.1- 0.9-CULT 0.1- 26.8* 61.5* 0.2- 2.1- 21.3* 2.2- 6.0-O3XCULT 1.9- 3.7- 2.0-0.7- 2.9- 7.3- 1.1- 1.5-O3 LINXCULT 4.0-O3 QUADXCULT 3.7- 20.5* 0.1- 6.8- 19.8* 1.4- 2.4-O3 DEVXCULT 1.8- 0.8- 0.1- 3.4* 0.2- 0.2- 1.4- 0.7-BW = Bud weight (g); FW = Flower weight (g); PW = Pod weight (g); MPW = Marketable pod weight (g); DLW = Dead leaf weight (g); TW = Total weight (g); SW = Stem weight (g); LW = Le a f l e t weight (g); HT = Plant height (cm); LN = Leaf number; DLN = Dead leaf number; BN = Bud number; FN = Flower number; PN = Pod number; LA = Leaf area (sq. cm); MPN = Marketable pod number. = ns * p = 0.05 ** p = 0.01 *** p = 0.001 60 Table 6. Summary of anova r e s u l t s : F-values for the e f f e c t s of ozone on carrot v a r i a b l e s (per plant) SOURCE OF VARIATION df LN RW LL RL LW TW °3 (3) O3 LIN 1 0 3 QUAD 1 O3 DEV 2 CULTIVAR 1 O3XCULT (3) O3 LINXCULT 1 O3 QUADXCULT 1 O3 DEVXCULT 2 1.04- 1.17-1.02-2.70- 4.26- 1.34-0.53- 10.12- 3.91-0.80- 1.04- 1.16-0.41- 1.12-1.48-0.94- 0.70- 0.86-10.28- 0.60- 0.87-0.60- 5.60- 4.96-16.38- 0.29- 5.73-1.69- 0.93- 1.04-3.07- 0.93- 1.13-0.42- 0.73- 0.67-LN = Leaf number; RW = Root dry weight (g); LL = Length of the longest leaf (cm); RL = Root length (cm); LW = Leaf weight (g); TW = Tot a l weight (g) . = ns * p = 0.05 ** p = 0.01 *** p = 0.001 Table 7. Simple l i n e a r regressions of b r o c c o l i inflorescence y i e l d on various exposure s t a t i s t i c s . ( Y i e l d = a + b[Exposure]) Exposure a b r 2 RMS p s t a t i s t i c (g) cv. Emperor: P1[12] 64 .6829 -0.0645 P1[14] 64 .6761 -0.0645 P7 83 .3174 -0.5345 P12 73.6281 -0.2654 Ml[7] 85 .9094 -0.8446 P14 73 .7111 -0.3072 M7 90 .0900 -1.2437 M14 80 .2861 -0.8386 M1[12] 69 .5045 -0.3055 M12 80 .4262 -0.8342 M1[14] 68 .5154 -0.2883 Pl[7] 74 .6628 -0.2981 0 .985 2 .362 0.007 0 .985 2 .374 0.007 0 .973 4 .395 0.014 0 .955 7 .180 0.023 0 .949 8 .244 0.026 0 .945 8 .097 0.026 0 .929 11 .421 0.036 0 .923 12 .440 0.039 0 .919 13 .009 0.041 0 .918 13 .231 0.042 0 .846 24 .800 0.080 0 .743 41 .315 0.138 cv. SGI P1[12] 99 .1279 P1[14] 99 .1202 Ml [7] 129 .9760 M7 136 .4269 M1[12] 106 .3106 M12 122 .0291 M14 121 .5573 P12 110 .5755 M1[14] 105 .2123 P7 123 .1599 P14 110 .1876 Pl[7] 110 .8646 -0.0933 0.990 -0.0933 0.990 -1.2253 0.957 -1.8151 0.948 -0.4450 0.934 -1.2103 0.926 -1.2103 0.921 -0.3700 0.890 -0.4241 0.878 -0.7321 0.875 -0.4228 0.862 -0.4038 0.654 0.335 0.005 3.325 0.005 14.424 0.022 17.328 0.026 22.044 0.033 24.820 0.038 26.484 0.040 36.912 0.057 41.096 0.063 42.105 0.065 46.286 0.072 116.319 0.192 Table 8. Simple l i n e a r regressions of bean pod y i e l d on various exposure s t a t i s t i c s . ( Y i e l d = a + b[Exposure]) Exposure a b r 2 RMS p s t a t i s t i c (g) cv. BBL-GV2 P1[12] 12 .5084 -0 .0098 0 .958 0 .136 0 .021 Pl[14] 12 .5084 -0 .0098 0 .958 0 .136 0 .021 P14 13 .4628 -0 .0444 0 .947 0 .173 0 .027 P12 13 .3016 -0 .0371 0 .929 0 .232 0 .036 P7 13 .9363 -0 .0613 0 .891 0 .352 0 .056 Pl[7] 13 .6388 -0 .0404 0 .861 0 .450 0 .072 M7 14 .6602 -0 .1394 0 .843 0 .510 0 .082 M1[12] 12 .4772 -0 .0386 0 .830 0 .552 0 .089 M14 13 .7117 -0 .1037 0 .825 0 .567 0 .092 M12 13 .7298 -0 .1009 0 .822 0 .577 0 .093 Ml[7] 14 .0058 -0 .0913 0 .814 0 .604 0 .098 Ml[14] 12 .0676 -0 .0312 0 .741 0 .840 0 .139 cv. Galamore P1'[14] 9. 7692 -0 .0062 Pl[12] 9. 7692 -0 .0062 P14 10. 3115 -0 .0274 P12 10. 1080 -0 .0219 M14 10. 0933 -0 .0550 M12 10. 0582 -0 .0525 M7 10. 4765 -0.0708 Ml[12] 9. 3512 -0 .0194 M1[14] 9. 1903 -0 .0162 Ml[7] 10. 0424 -0 .0443 P7 9. 8455 -0 .0276 Pl[7] 9. 1140 -0 .0125 0.388 1 .951 0 .377 0.388 1 .951 0 .377 0.366 2 .022 0 .396 0.328 2 .141 0 .427 0.236 2 .435 0 .514 0.226 2 .466 0 .525 0.221 2 .483 0 .530 0.213 2 .509 0 .539 0.203 2 .540 0 .550 0.195 2 .565 0 .558 0.183 2 .602 0 .572 0.084 2 .920 0 .711 Table 9. Simple l i n e a r regressions of pea pod y i e l d on various exposure s t a t i s t i c s . ( Y i e l d = a + bfexposure]) Exposure a b r 2 RMS p s t a t i s t i c (g) cv. Puget M1[14] 9.0401 -o-.0249 0.437 1 .966 0 .339 Ml[7] 10.2309 -0 .0654 0.395 2 .110 0 .371 M1[12] 9.0698 -0 .0271 0.377 2 .177 0 .386 M12 9.9424 -0 .0706 0.372 2 .194 0 .390 M14 9.8846 -0 .0716 0.363 2 .227 0 .398 M7 10.4855 -0 .0944 0.361 2 .234 0 .399 P7 9.7867 -0 .0389 0.332 2 .336 0 .424 Pl[7] 9.5037 -0 .0247 0.298 2 .452 0 .454 P12 8.9021 -0 .0189 0.224 2 .711 0 .527 P14 8.7775 -0 .0204 0.185 2 .847 0 .570 Pl[12] 8.2309 -0 .0042 0.161 2 .934 0 .599 P1[14] 8.2309 -0 .0042 0.161 2 .934 0 .599 cv. Bolero Pl[7] 8.7160 -0 .0180 0.399 0 .841 0 .368 P7 8.3318 -0 .0205 0.230 1 .077 0 .521 Ml[7] 8.2155 -0 .0276 0.176 1 .153 0 .581 M1[12] 7.7162 -0 .0113 0.164 1 .170 0 .595 M7 8.3200 -0 .0398 0.160 1 .17 5 0 .600 M12 8.0126 -0 .0279 0.145 1 .196 0 .619 M14 7.9712 -0 .0278 0.137 1 .208 0 .630 M1[14] 7.5608 -0 .0087 0.134 1 .212 0 .634 P12 7.6791 -0 .0082 0.105 1 .252 0 .676 P14 7.6278 -0 .0089 0.088 1 .277 0 .704 P1[12] 7.4005 -0 .0018 0.079 1 .289 0 .720 P1[14] 7.4005 -0 .0018 0.076 1 .289 0 .720 Table 10. Simple l i n e a r regressions of carrot root y i e l d on various exposure s t a t i s t i c s . ( Y i e l d = a + b[Exposure]) Exposure a b r 2 RMS p s t a t i s t i c (g) cv. Hipak P1[12] 6.4932 -0.0083 0 .428 2.978 0.346 P1[14] 6.4932 -0.0083 0 .428 2.978 0.346 P14 7.3094 -0.0377 0 .425 2.995 0.348 P12 7 .1464 -0.0313 0 .410 3.071 0.360 M14 7.3939 -0.0850 0 .397 3.141 0.370 M1[14] 6.4767 -0.0314 0 .394 3.154 0.372 M12 7.4070 -0.0827 0 .388 3.189 0.377 M7 7.9916 -0.1090 0 .368 3.289 0.393 Ml [7] 7.6299 -0.0738 0 .363 3.316 0.397 M1[12] 6.2849 -0.0302 0 .347 3.398 0.411 P7 6.9774 -0.0422 0 .263 3.840 0.488 Pl[7] 6.9293 -0.0283 0 .220 4.064 0.531 cv. Sixpak M1[14] 6 .6546 -0 .0190 0.263 2 .110 0.487 M14 7 .1406 -0 .0499 0.248 2 .153 0.502 M12 7 .1584 -0 .0487 0.245 2 .163 0.505 Ml[7] 7 .3086 -0 .0438 0.233 2 .196 0.517 M1[12] 6 .4913 -0 .0177 0.217 2 .241 0.534 M7 7 .3915 -0 .0613 0.212 2 .257 0.540 P12 6 .4112 -0 .0128 0.125 2 .506 0.647 P7 6 .5497 -0 .0201 0.108 2 .554 0.671 P14 6 .3385 -0 .0139 0.105 2 .562 0.676 P1[14] 5 .9588 -0 .0028 0.090 2 .607 0.701 P1[12] 5 .9588 -0 .0028 0.090 2 .607 0.701 Pl[7] 5 .8090 -0 .0069 0.024 2 .796 0.847 Figure 7: Relationship between yield (g) for Broccoli cv. Emperor and ozone exposure expressed as Pl[12] (ppb). BROCCOLI, CV. EMPEROR 67 Figure 8: Re la t ionsh ip between y i e l d (g) for B r o c c o l i cv . SGI and ozone exposure expressed as Pl[14] (ppb). BROCCOLI, CV. SG1 69 Figure 9: Relationship between yield (g) for Bean cv. BBL-GV2 and ozone exposure expressed as Pl[12] (ppb). 70 71 4.4.2 Polynomial Regression Analyses From the r e s u l t s of the anova for each crop i t i s c l e a r that for beans and peas i n p a r t i c u l a r the r e l a t i o n s h i p between y i e l d and ozone i s best explained by a quadratic function. The computed quadratic regressions for these crops and for b r o c c o l i and carrots are shown i n Tables 11 through 14. L i t t l e improvement i n f i t , as seen by the c o e f f i c i e n t s of determination, i s found for either b r o c c o l i c u l t i v a r s or bean cv. BBL-GV2 above that for a simple l i n e a r regression (Tables 7 and 8 vs 11 and 12). However for a l l other c u l t i v a r s there were improvements i n f i t although only i n the case of Galamore (bean), Bolero (pea) and Hipak (carrot) were these improvements s u f f i c i e n t to reach s i g n i f i c a n c e . Although for Emperor ( b r o c c o l i ) , Bolero (pea) and Hipak ( c a r r o t ) , a 1-h peak exposure s t a t i s t i c provided the best f i t , f o r SGI (broccoli) and Galamore (bean), the 14-h and 12-h peaks proved to be marginally best, r e s p e c t i v e l y . F i e l d observations indicated that there were p l o t to p l o t differences for s o i l f a c t o r s and disease i n the present study, and.that these d i f f e r e n c e s probably contributed to the growth responses observed. Although lack of true r e p l i c a t i o n had been recognized at the outset as an inadequacy of the experimental design, previous use of the ZAPS (Runeckles et a l . , 1981b) had not experienced the magnitudes of the observed inherent v a r i a t i o n among the blocks and the a d d i t i o n a l v a r i a t i o n r e s u l t i n g from disease problems that occurred i n the present experiment. Because of these inadequacies, and because resources were not a v a i l a b l e to increase the number of blocks within the ZAPS, the experimental design and the ZAPS d e l i v e r y system were changed for the experimentation conducted i n 1986. Table 11. Polynomial regressions of b r o c c o l i inflorescence y i e l d on various exposure s t a t i s t i c s . ( Y i e l d = a + [Exposure] + b 2[Exposure] 2) Exposure a b-^  b 2 r 2 RMS p s t a t i s t i c (g) cv. Emperor: Pl[7] 189 .1132 -2 .5853 0.010663 0 .999 0 .086 0.016 Pl[12] 71 .5198 -0 .1263 0.000107 0 .993 2 .379 0. 086 P1[14] 71 .5556 -0 .1266 0.000108 0 .993 2 .381 0. 086 P14 53 .6390 0 .2458 -0.003325 0 .985 4 .692 0. 121 P7 98 .0146 -0 .9823 0.003217 0 .978 7 .115 0. 149 P12 60 .2815 0 .0468 -0.001603 0 .971 9 .445 0. 171 Ml[7] 120 .5204 -2 .4001 0.016616 0 .962 12 .328 0. 196 M7 149 .4066 -4 .7392 0.049491 0 .956 14 .329 0. 211 M1[12] 83 .2311 -0 .7234 0.002743 0 .944 17 .950 0. 236 M12 107 .6143 -2 .2572 0.017427 0 .937 20 .258 0. 251 M14 100 .1275 -1 .8957 0.013149 0 .934 21 .337 0. 258 M1[14] 87 .9000 -0 .8514 0.003558 0 .911 28 .551 0. 298 cv. SGI: P14 53 .2605 1 .1455 -0 .009430 0. 999 0.056 0 .009 P1[14] 88 .9216 -0 .0011 -0 .000160 0. 998 1.445 0 .046 P1[12] 88 .8927 -0.0008 -0 .000161 0. 998 1.446 0 .046 P12 60 .5226 0 .8008 -0 .006010 0. 993 4.704 0 .084 Ml[7] 84 .1710 0 .8333 -0 .021990 0. 968 21.563 0 .179 Pl[7] 292 .8003 -4 .0396 0 .016950 0. 964 24.052 0 .189 M7 107 .2179 -0 .0938 -0 .024371 0. 952 32.591 0 .220 Ml[12] 98 .8418 -0 .2076 -0 .001492 0. 938 41.700 0 .249 M14 90 .5366 0 .4424 -0 .020557 0. 934 44.307 0 .257 M12 99 .5650 -0 .0345 -0 .014399 0. 932 45.404 0 .260 Ml [14] 109 .9812 -0 .5627 0 .000875 0. 880 80.918 0 .347 P7 128 .0350 -0 .8806 0 .001067 0. 875 84.026 0 .354 Table 12. Polynomial regressions of bean pod y i e l d on various exposure s t a t i s t i c s . ( Y i e l d = a + b-j_ [Exposure] + b 2 [Exposure] 2) Exposure a b± b 2 r 2 RMS f s t a t i s t i c (g) cv. BBL--GV2: P1[12] 12 .2660 -0 .0076 P1[14] 12 .2660 -0 .0076 P7 22 .0177 -0 .3030 P14 13 .6163 -0 .0486 Ml[7] 23 .9136 -0 .5178 Ml[l2] 15 .9109 -0 .1439 M7 25 .3370 -0 .7505 P12 14 .4819 -0 .0647 M12 20 .9407 -0 .4741 M1[14] 15 .3014 -0 .1235 M14 20 .4496 -0 .4655 Pl[7] 20 .8153 -0 .1874 0 .0000039 0. 959 0. 265 0 .202 0 .0000039 0. 959 0. 265 0 .202 0 .0016720 0. 955 0. 295 0 .213 0 .0000250 0. 947 0. 346 0 .231 0 .0042940 0. 946 0. 350 0 .232 0 .0006830 0. 939 0. 399 0 .248 0 .0082860 0. 936 0. 412 0 .252 0 .0001420 0. 935 0. 425 0 .256 0 .0044620 0. 930 0. 456 0 .265 0 .0005450 0. 925 0. 486 0 .274 0 .0044810 0. 925 0. 487 0 .274 0 .0006960 0. 881 0. 770 0 .345 cv. Galamore: P12 22 .5599 -0 .3131 0 .0014950 0. 999 0.003 0. 023 P7 38.6402 -0 .8887 0 .0051590 0. 998 0.014 0. 047 P14 22 .9690 -0 .3695 0 .0020370 0.988 0.077 0. 110 P1[12] 15 .5745 -0 .0593 0 .0000945 0. 969 0.197 0. 176 P1[14] 15 .5745 -0 .0593 0 .0000945 0. 969 0.197 0. 176 M7 39 .5535 -1 .7350 0.0225670 0. 928 0.461 0. 269 M14 27 .4882 -0.9890 0 .0115680 0. 914 0.550 0. 294 M1[12] 17 .9739 -0 .2837 0 .0017150 0. 910 0.574 0. 300 M12 28 .0361 -0 .9829 0 .0111250 0. 907 0.593 0. 305 Ml[7] 32 .4666 -1 .0096 0 .0097180 0. 884 0.740 0. 341 M1[14] 15 .1582 -0 .1864 0 .0010050 0. 841 1.016 0. 399 Pl[7] 34 .0557 -0 .5236 0 .0024180 0. 333 4.254 0. 817 Table 13. Polynomial regressions of pea pod y i e l d on various exposure s t a t i s t i c s . ( Y i e l d = a + ^[exposure] + b2texposure] 2) Exposure a -b^ b 2 r 2 RMS p s t a t i s t i c (g) cv. Puget; M12 -8 .0387 0.8602 -0 .0111360 0.997 0. 023 0 .058 Ml 4 -7 .6672 0.8725 -0 .0117140 0.996 0. 027 0 .062 Ml [7] -11 .5918 0.8716 -0 .0094000 0.995 0. 033 0 .069 Ml[12] 0 .5691 0.2341 -0 .0016990 0.991 0. 062 0 .094 Ml[14] 3.2363 0.1404 -0 .0009760 0.987 0. 094 0 .116 M7 -17 .9476 1.5293 -0 .0219640 0.982 0. 127 0 .135 P7 -12 .8453 0.6379 -0 .0046840 0.791 1. 465 0 .458 P12 -2 .8238 0.2553 -0 .0014080 0.767 1. 629 0 .483 P14 -2 .6943 0.2896 -0 .0018460 0.652 2. 435 0 .590 P1[14] 3 .2677 0.0413 -0 .0000808 0.548 3. 160 0 .672 P1[12] 3 .2677 0.0413 -0 .0000808 0.548 3. 160 0 .672 Pl[7] 24 .4319 -0.3306 0 .0014470 0.380 4. 337 0 .788 cv. Bolero: Pl[12] 2 .5578 0.0425 -0 .0000788 1.000 0.000 0 .004 P1[14] 2 .5578 0.0425 -0 .0000788 1.000 0.000 0 .004 P14 -2 .4520 0.2635 -0 .0016220 0.987 0.037 0 .115 P12 -1 .5160 0.2069 -0 .0011040 0.939 0.171 0.247 P7 -9 .2049 0.5040 -0 .0036290 0.918 0.229 0 .286 M7 -7 .5339 0.8657 -0 .0122480 0.642 1.001 0 .598 Pl[7] 6 .4656 0.2931 -0 .0014720 0.609 1.094 0 .625 M1[12] 3 .1794 0.1281 -0 .0009070 0.601 1.117 0 .632 Ml [7] -3 .0291 0.4552 -0 .0048430 0.573 1.194 0 .653 M14 -1 .2289 0.4670 -0.0061400 0.572 1.199 0 .655 M12 -1 .3289 0.4557 -0 .0057850 0.566 1.216 0 .656 M1[14] 5 .0579 0.0626 -0 .0004210 0.389 1.709 0 .782 Table 14. Polynomial regressions of carrot root y i e l d on various exposure s t a t i s t i c s . (Yield = a + ^[Exposure] + b 2 [Exposure] 2) Exposure a b^ b 2 r 2 RMS p s t a t i s t i c (g) cv. Hipak: Pl[7] 43.0356 -0 .7531 0 .003406 0 .999 0 .007 0 .025 P1[14] 13.1966 -0 .0697 0 .000109 0 .903 1 .016 0 .312 P1[12] 13.1966 -0 .0697 0.000109 0 .903 1 .016 0 .312 P14 20.9286 -0 .4058 0 .002191 0 .866 1 .398 0 .366 P12 19.4908 -0 .3200 0 .001482 0 .814 1 .938 0 .431 P7 35.6895 -0 .9008 0 .005942 0 .758 2 .519 0 .492 M7 26.6753 -1 .1815 0 .014475 0 .613 4 .026 0 .622 M14 17.8626 -0 .6513 0 .006985 0 .606 4 .107 0 .628 M12 18.4735 -0 .6596 0 .006882 0 .599 4 .181 0 .634 M1[14] 11.5450 -0 .1798 0 .000913 0 .591 4 .259 0 .640 Ml[7] 22.1497 -0 .6951 0 .006187 0 .581 4 .368 0 .648 M1[12] 12.2480 -0 .2123 0 .001164 0 .579 4 .385 0 .649 cv. Sixpak: M1[14] 2.9638 0 .0890 -0.0006650 0. 453 3.133 0 .740 M14 0.0895 0 .3316 -0.0047050 0. 420 3.319 0 .761 M12 -0.2403 0 .3370 -0.0046010 0. 416 3.344 0 .764 Pl[7] 24.6665 -0 .3854 0.0017787 0. 414 3.377 0 .768 Ml[7] -2.1280 0 .3600 -0.0040210 0. 400 3.436 0 .775 M1[12] 2.9024 0 .0919 -0.0007004 0. 370 3.608 0 .794 M7 -2.8218 0 .5250 -0.0079120 0. 345 3.751 0 .809 P12 5.7866 0 .0018 -0.0000750 0. 127 5.000 0 .934 P1[14] 7 .1370 -0 .0136 0.0000192 0. 116 5.060 0 .940 Pl[12] 7.1370 -0 .0136 0.0000192 0. 116 5.060 0 .940 P7 5.2427 0 .0190 -0.0002705 0. 110 5.098 0 .944 P14 7.3663 -0 .0417 0.0001654 0. 110 5.098 0 .944 76 5. MATERIALS AND METHODS-1986 EXPERIMENT. 5.1 EXPERIMENTAL DESIGN The 1986 experiment was conducted on the same s i t e as i n 1985. The ozone treatments were again arranged i n three blocks with the fourth block serving as an ambient a i r con t r o l as generally depicted i n Figure 2. However, for 1986 each 10 by 12 metre block was subdivided i n t o four 5 x 6 m p l o t s . Within each p l o t , four adjacent rows of peas and four of potatoes were randomly established. There was a 1-m wide border surrounding each p l o t leaving a t o t a l experimental area of 3 by 4 m per treatment (Figure 10). Plants were not harvested from border rows. 5.2 FIELD PLOTS AND GAS DELIVERY SYSTEM The basic gas d e l i v e r y system used was the same as that described previously. However, the same ozone-enriched a i r was supplied to a l l three manifolds, but d i f f e r e n t amounts were released over the d i f f e r e n t p l o t s by varying the numbers of o r i f i c e s i n the manifolds above the p l o t s . In each block, one p l o t received ozone-enriched a i r from 36 h o r i z o n t a l l y opposed o r i f i c e s (spaced at lm i n t e r v a l s ) while the others received a i r discharged from 72, 108, or 144 o r i f i c e s . In these l a t t e r cases the o r i f i c e s were grouped together i n 2 1s 3 1s or 4's (1 cm apart) at 1-m spacing. The locations of the p l o t s with the d i f f e r e n t numbers of o r i f i c e s were randomly selected for each block (Figure 11). The close proximity of the p l o t s within each block resulted i n random carry-over from one p l o t to another, depending upon wind speed and d i r e c t i o n and the geographic r e l a t i o n s h i p s of the pl o t s to each other. The consequence of these modifications was to provide a t o t a l of 12 p l o t s within the three blocks to which supplementary 77 Figure 10: Network of a d i s t r i b u t i o n manifold of the ZAPS at the U n i v e r s i t y of B r i t i s h Columbia. 78 SAMPLING LOCATION SUPPLY LINE ORIFICE(S) SUB-PLOT FOR OZONE STUDIES 79 Figure 11: Layout of field experimental plots at the University of British Columbia. - Numbers indicate the frequency of discharge orifices over the plots. 80 BLOCK 2 BLOCK 4 4 1 2 5 M 1 2 2 3 i k S u p p l y & 2 5 M I n s t r u m e n t a t i o n ' * r 12M ** 3 2 1 4 r 4 3 1 2 r BLOCK 1 • BLOCK3 81 ozone could be applied. Furthermore, the random e f f e c t s of wind were expected to r e s u l t i n uniquely d i f f e r e n t d i s t r i b u t i o n s of ozone concentrations on each of the 12 p l o t s , although the three d i f f e r e n t p l o t s with the same numbers of o r i f i c e s would be expected to have s i m i l a r (though not i d e n t i c a l ) long-term average concentrations. Two p l o t s were established on the control block making a t o t a l of 14 p l o t s . The ozone generator was i n i t i a l l y supplied with compressed oxygen. On August 8th the source of oxygen was switched to compressed a i r , supplied by an ITT Pneumotive o i l - l e s s a i r compressor (Model GH 7103-H30) o u t f i t t e d with a Van A i r MMS microdryer. Supply of ozone extended over a 14 hour period, 0700-2100 hours ( P a c i f i c daylight time (PDT)) commencing June 2nd (24 days from planting) and terminating August 17th. 5.3 OZONE MONITORING AND CONTROL The ozone monitor, Model 1003-AH (Dasibi Environmental Corporation, Glendale, CA), operated on a time-sharing basis among the 14 treatment p l o t s . A i r from each of the 12 treatment and two control p l o t s was drawn continuously through 1/4 inch t e f l o n l i n e s of equal length (30 m), each f i t t e d with a dust f i l t e r (Schleicher and Schuell Inc.,Keene, N.H., Grade TE37, 25 mm diam., 1.0 micron). The f i l t e r s were replaced every two weeks. Line losses were measured and found to be 0.1 ppb per metre. A manifold of solenoid valves sequentially d i r e c t e d a i r from each sample l i n e to the monitor for a 2-minute sampling period once every 30 minutes. Each 2-minute sampling period resulted i n the c o l l e c t i o n of s i x monitor readings. A l l solenoids were co n t r o l l e d by the time clock i n the ozone monitor. 82 The i n l e t s to the sampling l i n e s were located at potato plant canopy-height, and hence varied throughout the season. The ozone concentration released over the treatment pl o t s was again c o n t r o l l e d by a feedback program from the Campbell S c i e n t i f i c Model 21-X data a c q u i s i t i o n system used to record monitoring and other data. This feedback regulated the input voltage to the ozone generator. An algorithm for ozone generation, u t i l i z i n g information from the average of the s i x readings taken over two minutes every 30 minutes f o r the ambient a i r p l o t (AA2), was used to adjust the output from the servomotor c o n t r o l l i n g the generator. Ozone concentrations i n either treatments 3B1 or 4B1 ( i . e . the 3-and 4-hole p l o t s i n block 1, see Figure 11) were checked i n each c y c l e . If l e v e l s greater than 250 ppb were detected, the generator was shut down u n t i l subsequent readings le s s than 250 ppb were obtained from these l o c a t i o n s . Because of the tendency (observed i n 1985) f o r the supplementary ozone to r e s u l t i n higher than desired l e v e l s at the beginning and end of the d a i l y enrichment period, an a d d i t i o n a l control was b u i l t i n t o the program, based upon windspeed, measured using a Model W102-P Skyvane 1 wind sensor and s i g n a l conditioning t r a n s l a t o r (WeatherMeasure Corp., CA). This modified the output voltage supplied to the ozone generator i n a l i n e a r fashion, from zero gain at zero windspeed to f u l l gain at a windspeed of 10 m s - 1 or greater. The windspeed readings used by the program were moving averages of readings taken every two minutes over a 15-minute period. This modification was introduced on J u l y 16, approximately 44 days from the s t a r t of fumigation. Since the ozone generator was a high voltage d i e l e c t r i c generator and the source of oxygen l a t e i n the season was switched to compressed a i r , the 83 p o s s i b i l i t y of n i t r i c oxide formation was investigated. This was accomplished by monitoring the a i r i n 3B2 (3-hole p l o t i n block 2) for nitrogen oxide (NO) and nitrogen dioxide (N0 2) using a Thermo Electron (Model 14 D/E) Chemiluminescent NOX, NO, N02 analyzer, when ozone was generated from oxygen or from compressed a i r . Averages of 0.0042 ppm NO and 0.034 ppm N0 2 were observed p r i o r to i n s t a l l a t i o n of the a i r compressor and 0.0024 ppm NO and 0.015 ppm N0 2 a f t e r , i n d i c a t i n g that, i f nitrogen oxides were being generated and released, they contributed n e g l i g i b l y to the ambient l e v e l s . A d d i t i o n a l channels i n the Campbell 21-X c o l l e c t e d wind speed and wind d i r e c t i o n data, and nitrogen oxide and nitrogen dioxide measurements. A l l data were recorded every two minutes. As i n 1985, v e r t i c a l p r o f i l e s were obtained for two locations i n one d i s t r i b u t i o n manifold. The h o r i z o n t a l d i s t r i b u t i o n of ozone concentrations was determined u t i l i z i n g the sampling pattern shown i n Figure 12. Each of the 32 sampling points was monitored f o r a period of 15 minutes at a height of 40 cm above ground l e v e l , the measurements being taken during periods i n which, for each set of 16 points, the windspeeds and d i r e c t i o n were reasonably steady during the four hour period required to complete the data c o l l e c t i o n . Both v e r t i c a l and hori z o n t a l p r o f i l e s were measured over bare s o i l . 5.4 DATA HANDLING As i n 1985, a l l signals were passed to the Campbell S c i e n t i f i c Model 21X data a c c q u i s i t i o n system for i n i t i a l processing and were stored on cassette tapes; the data were subsequently transferred v i a a C20 cassette i n t e r f a c e to the U.B.C. mainframe computer for a n a l y s i s . 84 Figure 12: Sampling s i t e s f o r the h o r i z o n t a l p r o f i l e s measured i n block 2. Numbers represent p o s i t i o n s sampled for a 15 minute period within the manifold, at a height of 40 cm. 86 5.5 FIELD PREPARATION In e a r l y A p r i l , Roundup was applied at the recommended rates to k i l l a l l perennial weeds and grasses. After a two week period (allowing time for the herbicide to take e f f e c t ) the land was prepared by r o t o t i l l i n g followed by harrowing. On May 1 an a l l purpose f e r t i l i z e r (13:16:10) was broadcast at the rate of 900 kg/ha on the pl o t s and r o t o t i l l e d i n t o the s o i l . On May 5 the p l o t s were harrowed again and then raked by hand. The zonal a i r p o l l u t i o n system (ZAPS) was set up over the three treatment blocks May 7-8, but treatments d i d not commence u n t i l June 2, 24 days a f t e r p lanting. 5.6 PLANT MATERIALS The two plant species used i n t h i s study were pea (Pisum sativum .1. cv. Puget) and potato (Solanum tuberosum L. cv. Russett Burbank). These crops were chosen based on vegetable production data from the l a s t six years (data provided by M.K. James, Mi n i s t r y of A g r i c u l t u r e and Food, Surrey) which demonstrated t h e i r importance both i n terms of d o l l a r value and the amount of production i n the Lower Mainland. The pea seeds used were supplied by Royal C i t y Foods Ltd. from W. Brotherton Seed Co., Inc., Moses Lake, Washington, USA. The seed potatoes were supplied by A. Van Loon, Pemberton, B.C.. Seeds inoculated with legume Rhizobium (Nitragin C. Milwaukee, Wis.) were sown 5 cm deep i n rows 50 cm apart, May 9-10, 1986. Thinning to 5 cm apart took place when the seedlings had formed true leaves. The seed potatoes were cut i n t o approximately 60g pieces and coated with Captan. They were l e f t at room temperature for two days p r i o r to 87 p l a n t i n g . The potato pieces were sown 10 cm deep, 25 cm apart i n rows 1 m apart, May 9-10, 1986. The seeding rates for both crops were i n accordance with recomendations ou t l i n e d i n the 1986 Vegetable Production Guide for B.C. The growing seasons for both crops are summarized i n Table 15. Table 15. Crops and harvest dates (1986 season) Crop Harvest J u l i a n Growing Exposure date day season season Pea J u l y 31 211 83 58 Potato August 17, 18 228/9 100 75, 76 5.7 CROP PEST AND RODENT MANAGEMENT The p l o t s were examined d a i l y for pests, diseases, ozone i n j u r y symptoms and plant development during the experimental period. The peas were f u l l y emerged by May 19, 10 days from seeding. Potato shoots started to emerge on May 23 and were completely emerged by May 30, 21 days from p l a n t i n g . After planting, weeds were c o n t r o l l e d by hoeing and hand weeding between pl a n t s . Potatoes were h i l l e d on June 25, 46 days from planting. Tensiometers ( S o i l Moisture Equipment Corp. Santa Barbara, CA.) were i n s t a l l e d on May 22 at 10 and 15 cm depths i n each block. Though r a i n f a l l was prevalent throughout the growing period, supplementary i r r i g a t i o n by overhead s p r i n k l e r was necessary on June 11, August 1 and August 8. Due to 88 a broken water main, blocks 1 and 2 were flooded on June 3rd. To balance t h i s gain i n water, blocks 3 and 4 were i r r i g a t e d to f i e l d capacity June 4. Cutworms and aphids were observed i n a l l p l o t s during July. To con t r o l these pests, Diazinon was applied to the f o l i a g e on July 8 at the recomended r a t e s . By July 9 spray damage (approximately 10 % of the leaf area) had developed on the pea plants. Damage was more extensive i n most treatment p l o t s than i n the cont r o l p l o t s . An i n f e s t a t i o n of Early Bl i g h t ( A l t e r n a r i a solani) was observed on the potato plants by August 10. This i n f e c t i o n was worse on control and block 2 plants than on other treated plants, and spread so r a p i d l y that the de c i s i o n was made to terminate the experiment 20 days e a r l i e r than planned. Rodents were p a r t i c u l a r l y prevalent during July and August. In an attempt to discourage them from digging tunnels around the potato plants and feeding on tubers and pea plants, poison b a i t and traps were l a i d out i n the p l o t s . However these methods were only moderately successful. Flecking and bronzing of the potato leaves, a symptom of ozone i n j u r y , were observed during flower development i n early J u l y . Though the i n j u r y symptoms became more prevalant as the plants matured, they d i d not develop on a l l leaves of a plant nor on a l l plants i n a treatment. No attempt was made to quantify the percentage of f o l i a r i n j u r y per plant. By l a t e July the lower leaves of many potato plants had begun to senesce and t h e i r stems began to lodge. 89 5.8 HARVEST SCHEDULE AND PLANT MEASUREMENTS 5.8.1 Pea At harvest (July 31, 1986, 83 days from seeding) eight plants were randomly selected from the row and cut at ground l e v e l . Both pod and pea fresh weight were measured. Pea data were recorded on a per plant b a s i s . 5.8.2 Potato At harvest (August 17 and 18, 1986, 100 days from planting) six plants were randomly selected from the rows and the tops were cut at ground l e v e l . Tubers were l e f t i n the ground f o r 1 week to mature before being harvested. Tuber fresh weights were recorded per plant. 5.9 DATA ANALYSIS 5.9.1 Ozone d i s t r i b u t i o n To e s t a b l i s h the nature of the d i s t r i b u t i o n of ozone concentrations achieved for each treatment over the season, 2-minute averages (0900-2059) were grouped i n t o frequency classes and p l o t t e d . Normal and log-normal d i s t r i b u t i o n s were investigated. In addition, data from one of the ambient p l o t s were examined by the Maxfit program (Holland and Fitz-Simons, 1982) which tests f o r goodness of f i t to 3-parameter log-normal, 3-parameter gamma, 3-parameter Weibull and Johnson S B (4-parameter log-normal) d i s t r i b u t i o n s . D i f f e r e n t averaging times (2-min or 1-h) were used for d i f f e r e n t d a i l y time periods. 5.9.2 Exposure s t a t i s t i c s In a d d i t i o n to the exposure s t a t i s t i c s computed i n 1985, various exposure s t a t i s t i c s such as the sums of concentrations above various 90 thresholds, the sums of episodes above these thresholds, the sums of a l l concentrations, the geometric means, the mean of a l l hourly 1000-, 1200-and 1400-h (computed s t a r t i n g at the begining of the stated time) readings f o r the season; and the geometric means of these terms were derived from the ozone concentration data f o r each treatment p l o t and used i n the regression analyses. These s t a t i s t i c s are summarized i n Table 16. A l l exposure s t a t i s t i c s were based upon hourly means computed from the o r i g i n a l 2-minute means recorded, for the appropriate daytime period for the duration of the fumigations, for peas and potatoes separately (see Table 15). The D25 exposure s t a t i s t i c was also determined over d i f f e r e n t periods i n the season to take i n t o account the phenological development of the crops. 5.9.3 Regression analyses Simple l i n e a r and polynomial regression analyses were used to define the f u n c t i o n a l r e l a t i o n s h i p s among the fr e s h weight y i e l d s of pods, peas and potato tubers and the exposure s t a t i s t i c s derived. The regression curves were calculated from the treatment means since the plants per se were not true r e p l i c a t e s of the treatment, but simply provided an i n d i c a t i o n of the v a r i a t i o n within a treatment (Steel and T o r r i e , 1980). Stepwise multiple regression analysis was also used to evaluate the r e l a t i v e importance of several d i f f e r e n t expressions of exposure i n causing y i e l d responses. Table 16. Exposure s t a t i s t i c s used as independent v a r i a b l e s i n l i n e a r and polynomial exposure-response models. Except where noted, a l l s t a t i s t i c s r e fer to values obtained during the d a i l y 12-hour period i n which ozone was released, 0900-2059 h. Exposure Description Reference s t a t i s t i c M7 M12 Ml [7] Ml [12] P7 P12 P l [7] P l [12] SUM50; SUM80; SUM100. Season-long mean of d a i l y 7-h mean (0900-1559 h) Season-long mean of d a i l y 12-h mean Season-long mean of d a i l y 1-h max. (0900-1559 h) Season-long mean of d a i l y 1-h max. Peak 7-h season-long (0900-1559 h) mean Peak 12-h mean Single peak 1-h mean (0900-1559 h) for the season Single peak 1-h mean for the season Sums of the f r a c t i o n s of a l l ozone concentrations above a Heck et a l . (1984) Heagle et a l . (1987) Heck et a l . (1984) Heck et a l . (1984) Heck et a l . (1984) Oshima et a l . threshold of 50, 80 or 100 ppb (1976) IE50; IE80; IE100. Sums of the absolute ozone Lefohn & HGT80; HGT100. SUMALL D25; D50; D80. 2C25; 3C25; 4C25. GM12 H10; H12; H1.4. GH10; GH12; GH14. concentrations when thresholds Benedict of 50, 80 or 100 ppb (1982) were exceeded. Sums of the numbers of Lee et a l . hourly ozone concentrations (1987) above a threshold of 80 or 100 ppb Sum of a l l hourly means, Foster et over 24-h,for the season a l . (1983) Sums of the numbers of days where an hourly ozone concentration exceeded 25, 50 or 80 ppb Sum of the number of days with 2, -3 or 4 consecutive hours greater than 25 ppb ozone Season-long geometric mean of d a i l y 12-h geometric mean Season-long hourly means for hours 1000, 1200 or 1400 Season-long geometric hourly means for hours 1000, 1200 or 1400 92 6. RESULTS-1986 EXPERIMENT 6.1 AIR QUALITY 6.1.1 S p a t i a l d i s t r i b u t i o n s Ozone was again found to be reasonably uniformly d i s t r i b u t e d throughout the v e r t i c a l p r o f i l e down to the lowest point measured, as shown i n Figure 13. Minor differences i n the mean ozone concentrations with height above ground are evident and v a r i a t i o n i n concentration (as indicated by the standard deviation) at a l l heights i s apparent. Means of the more than 45 ozone concentrations measured during 15 minutes for each ho r i z o n t a l p o s i t i o n over the two locations sampled within a p l o t (Figure 12) are presented i n Table 17. The data i n d i c a t e that the arithmetic mean ozone concentrations tend to be lower at p o s i t i o n s close to the outer portions of the arms for both locations (positions 1, 5, 9 and 13 for each l o c a t i o n ) . This suggests a mild concentration gradient within each p l o t . At the time at which the measurements were made i n l o c a t i o n 1, the predominant wind d i r e c t i o n was from the south-east. It was from the south-west for l o c a t i o n 2. P r e v a i l i n g wind d i r e c t i o n may thus have contributed to these gradients. Though the grand means for locations one and two are comparable (19.54 ppb and 19.57 ppb r e s p e c t i v e l y ) , the c o e f f i c i e n t s of v a r i a t i o n i n d i c a t e that the concentrations were somewhat more v a r i a b l e at the outer edges of the block (38.79% versus 30.66% r e s p e c t i v e l y ) . These features of the ZAPS d i s t r i b u t i o n were a n t i c i p a t e d and plants were not harvested from the buffer zone. The data presented i n Figure 13 and Table 17 suggest that the supplementary ozone provided by the ZAPS was generally uniformly d i s t r i b u t e d over the indvidual p l o t s . 93 Figure 13: Vertical profile of mean ozone concentrations averaged over 2 locations (+ standard deviation), in block 2. —> = manifold height (cm). 120-1 , Q , Ld O TD GO o GO 100-80 6 0 -> o CD < 40-t— X CD UJ 20 H 10 O-• • • • • • • • O • • • • • • • • • • • • • • • • • • T " 15 T I 20 25 30 35 OZONE CONCENTRATION [ppb] Table 17. Horizontal d i s t r i b u t i o n of ozone i n 2 locations of the d i s t r i b u t i o n manifold. Mean O3 cone, (ppb) (1) (2) (3) (4) Location 1 11.32 16.51 18.97 19.15 (5) (6) (7) (8) 7.59 15.06 22.65 25.14 - (9) (10) (11) (12) 12.95 21.68 21.46 22.39 (13) (14) (15) (16) 17.23 25.66 27.26 27.84 Mean and CV. for l o c a t i o n 1 19.54 38.79% Location 2 (1) (2) (3) (4) 13.94 18.29 21.29 20.10 (5) (6) (7) (8) 12.35 20.23 21.08 18.25 (9) (10) ( I D (12) 13.29 18.44 20.84 21.60 (13) (14) (15) (16) 23.22 20.31 22.10 27.40 Mean and CV. 19.57 30.66% for l o c a t i o n 2 NB. Numbers i n brackets are po s i t i o n s sampled within the d i s t r i b u t i o n manifold (see Figure 12). 96 6.1.2 Temporal d i s t r i b u t i o n s The season-long patterns of the average ozone concentrations achieved throughout the day are shown i n Figures 14 through 16. Because of the random assignment of p l o t s , the season-long mean concentrations observed d i d not n e c e s s a r i l y follow the same order as the number of o r i f i c e s d i c t a t e d , because of carry-over e f f e c t s . Thus i n block 1 (Figure 14), the season-long 1-hole p l o t and 3-hole p l o t mean concentrations were coincident and lower than the 2-hole p l o t . In block 2 (Figure 15) the 3-hole and 4-hole p l o t mean concentrations were coincident for much of the day whereas i n block 3 (Figure 16) the 2-hole and 4-hole p l o t mean concentrations were si m i l a r and lower than the 3-hole p l o t . As with 1985, p e r f e c t l y proportional additions of ozone were not achieved. However, the d i u r n a l pattern of 0 3 concentrations, on a seasonal basis, c l e a r l y resembled that of ambient a i r . It should be pointed out that the data presented here are season-long means for each hour and as such do not n e c e s s a r i l y r e f l e c t the s i t u a t i o n that may have occurred on any one day. The range of the mean seasonal 12-h ozone concentrations (M12) achieved during the experimental period for peas and potatoes, across a l l treatments was 13 to 42 ppb (Appendix 3). There was an approximately 3 ppb d i f f e r e n c e between the two ambient a i r p l o t s that p e r s i s t e d throughout the season (Figure 14). This d i f f e r e n c e apparently re s u l t e d from a combination of systematic differences i n O3 i n the sampling l i n e s and sampling manifold. For example, when the i n l e t s of both sample l i n e s from the ambient p l o t s were bought together, and a t h i r d l i n e run i n p a r a l l e l to a second monitor, the range of ambient ozone concentrations observed was plus or minus 1.5 ppb. 97 F igure 14: Hourly seasonal mean ozone concentrat ions for block 1, treatments 1 to 4 and AA 1 and 2 for the per iod of June 2 to August 18, 1986. 70-, A AMBIENT AIR 1 X AMBIENT AIR 2 • . TREATMENT 1 BLOCK 1 TREATMENT 2 BLOCK 1 TREATMENT 3 BLOCK 1 X- TREATMENT 4 BLOCK 1 99 F igure 15: Hourly seasonal mean ozone concentrat ions for block 2, treatments 1 to 4 and AA 1 and 2 for the p e r i o d of June 2 to August 18, 1986. 70 -i A AMBIENT AIR 1 X AMBIENT AIR 2 • TREATMENT 1 BLOCK 2 13 TREATMENT 2 BLOCK 2 XX TREATMENT 3 BLOCK 2 X TREATMENT 4 BLOCK 2 6 0 -° i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 212223 24 TIME hr 101 Figure 16: Hourly seasonal mean ozone concentrations f or block 3, treatments 1 to 4 and AA 1 and 2 for the period of June 2 to August 18, 1986. 70-. A AMBIENT AIR 1 X AMBIENT AIR 2 • TREATMENT 1 BLOCK 3 TREATMENT 2 BLOCK 3 TREATMENT 3 BLOCK 3 X- TREATMENT 4 BLOCK 3 6 0 -Q_ C L "1 1 1 1 1 1—I 1 1 1 I—I 1 1 1 1 1—I—I 1 1 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 2122 23 24 TIME hr 103 The sharp peaks occurring l a t e r i n the day (2000 h) i n some seasonal treatment means (Figures 14 to 16) were l i k e l y due to the system's i n a b i l t y to respond f a s t enough as the ambient concentration decreased. Although the mechanical problem noted i n 1985 had been corrected, i t was noted that windspeed tended to drop at these times, and t h i s allowed buildup of O3 around the p l a n t s . The windspeed c o r r e c t i o n , described above (Section 5.3), remedied t h i s s i t u a t i o n , but was not implemented u n t i l July 16th, i . e . 44 days a f t e r the st a r t ' o f the fumigation, and i t s e f f e c t was i n s u f f i c i e n t to overcome the higher than desired concentrations which had occurred e a r l i e r . One of the concerns expressed with the use of a ZAPS i s the adequacy of the monitoring (Krupa, 1984). Frequent measurements i n a number of locations are recommended i n order to define adequately the exposure that the plants receive. The system used employed only a sin g l e time-shared monitor and a sin g l e sampling l o c a t i o n per p l o t . As a r e s u l t , each p l o t could only be sampled once i n a 30 minute period. This was d i c t a t e d by the need to ensure that six sequential readings could be taken for each sample at the appoximately 20 second frequency of the Dasibi monitor. In order to estimate whether or not the concentrations recorded were t r u l y representative of the exposures, an a d d i t i o n a l sample l i n e was i n s t a l l e d on p l o t three, block two, with i t s i n l e t immediately adjacent to the e x i s t i n g sample l i n e , connected to a second O3 monitor. The two monitors were run for 24-h/day from May 12 to 18, 1987 while the system was operating. The frequency d i s t r i b u t i o n s f o r these two data sets showed some di f f e r e n c e s : one monitor and sample l i n e had fewer observations i n the 11-20 ppb c l a s s than the other and more i n the higher frequency cla s s e s . However, there was l i t t l e consistent r e l a t i o n s h i p between the two records. 104 On several occasions a high reading from one monitor was not r e f l e c t e d i n the reading from the other, while on other occasions both monitors responded s i m i l a r l y , suggesting that the measurements recorded were " r e a l " and not a r t i f a c t s of the instruments or systems. Although these observations were made more frequently than the 30-minute cycles used i n 1986, the actual d i s t r i b u t i o n s achieved resembled those of the 1986 data (see below) with few values i n the high frequency classes, suggesting that the 30-minute monitoring c y c l e provided a v a l i d p i c t u r e of a i r over the p l o t s . With regard to the concentrations measured over the i n d i v i d u a l p l o t s , i t i s clear from Figures 17 through 20 that the ozone concentrations (2-min average, 0900-2059) achieved over the season were not normally d i s t r i b u t e d . Figures 21 through 24, showing cumulative frequency p l o t s of the logarithms of these ozone concentrations suggest that the data are better described i n terms of a log-normal than a normal d i s t r i b u t i o n , although they c l e a r l y deviate from the l i n e a r form required for a log-normal d i s t r i b u t i o n . Analysis of the d i s t r i b u t i o n s of concentrations i n AA1 for goodness of f i t with respect to six t h e o r e t i c a l d i s t r i b u t i o n s , using the MAXFIT program (Holland and Fitz-Simons, 1982), showed that, when a 2-minute averaging time was used for the 14-h (0700-2059) enrichment period, the data were best described by a log-normal d i s t r i b u t i o n (Appendix 4, Table A4-3). If the span of time was increased from 14-h to 24-h (Appendix 4, Table A4-6) the best f i t was found with a Weibull d i s t r i b u t i o n . A change i n the averaging time from two minutes to one hour, for 24 hours resulted i n best f i t s with either a Johnson S B (4-parameter log-normal d i s t r i b u t i o n ) or a gamma (3-parameter) d i s t r i b u t i o n , depending on which s e l e c t i o n 105 Figure 17: D i s t r i b u t i o n s of ozone concentrat ions (2 minute average between 0900-2059, June 2 to August 18) for the ambient a i r p l o t s , 1986. AMBIENT AIR 1 8OO-1 600-400-200-l 1 1 1 r — i 1 1 — 5 55 105 155 205 255 305 355 405 OZONE CONCENTRATION [ppb] AMBIENT AIR 2 8OO-1 600-400-O U l or 200-T T 55 105 155 — i 1 1 r — i — 205 255 305 355 405 OZONE CONCENTRATION [ppb] o CM 107 Figure 18: D i s t r i b u t i o n s of ozone concentrations (2 minute average between 0900-2059, June 2 to August 18) for block 1, treatments 1 to 4, 1986. 108 TREATMENT 1 BLOCK 1 800 00 z o ac Ul 00 m O >-o UJ O 600-400 QC 200-I I I 1 1 I 5 55 105 155 205 255 305 355 405 OZONE CONCENTRATION [ppb] TREATMENT 2 BLOCK 1 to z o I cc Ul lO CO o 800 600 u. *00 >-o 3 O or 200 5 55 105 155 205 255 305 355 405 OZONE CONCENTRATION [ppb] TREATMENT 3 BLOCK to U l 00 m O >-3 s cc u 800 600 400-200 55 105 155 205 255 305 355 405 OZONE CONCENTRATION [ppb] TREATMENT 4 BLOCK 1 800 600 2 a: ui 00 m O >-o o UJ or 400 200 i 1 1 r 155 205 255 305 355 405 5 55 OZONE CONCENTRATION [ppb] 109 Figure 19: D i s t r i b u t i o n s of ozone concentrations (2 minute average between 0900-2059, June 2 to August 18) for block 2, treatments 1 to 4, 1986. 110 TREATMENT 1 BLOCK 2 TREATMENT 2 BLOCK 2 in z g or Ld to ca O >-o ZD o 800 600-400 or 200 800 ' i " " T i 1 r — i r 5 55 105 155 205 255 305 355 405 OZONE CONCENTRATION [ppb] T 1" T " I I I 55 105 155 205 255 305 355 405 OZONE CONCENTRATION [ppb] TREATMENT 3 BLOCK 2 TREATMENT 4 BLOCK 2 800 800 5 55 105 155 205 255 305 355 405 OZONE CONCENTRATION [ppb] to z g cc Ld tn CO O 600 U_ 400 >-o 3 O or 200 5 55 105 155 205 255 305 355 405 OZONE CONCENTRATION [ppb] I l l Figure 20: D i s t r i b u t i o n s of ozone concentrations (2 minute average between 0900-2059, June 2 to August 18) for block 3, treatments 1 to 4, 1986. TREATMENT 1 BLOCK 3 112 TREATMENT 2 BLOCK 3 oo z o ac U l to m O 800 600 U 4 0 0 i >-o => o Ct 200 800 5 55 105 155 205 255 305 355 405 OZONE CONCENTRATION [ppb] 5 55 105 155 205 255 305 355 405 OZONE CONCENTRATION [ppb] TREATMENT 3 BLOCK 3 TREATMENT 4 BLOCK 3 800 800 T 1 r — i r 5 55 105 155 205 255 305 355 405 OZONE CONCENTRATION [ppb] i i i r 5 55 105 155 205 255 305 355 405 • OZONE CONCENTRATION [ppb] 113 Figure 21: Cumulative d i s t r b u t i o n s of ozone concentrations (ppb) (0900 to 2059 h) for the ambient a i r p l o t s , block 4, plo t t e d on l o g - p r o b a b i l i t y paper. 115 F igure 22: Cumulative d i s t r i b u t i o n s of ozone concentrat ions (ppb) (0900 to 2059 h) for treatments 1 to 4, b lock 1, p l o t t e d on l o g - p r o b a b i l i t y paper. 116 CUMULATIVE PROBABILITY 117 F igure 23: Cumulative d i s t r i b u t i o n s of ozone concentrat ions (ppb) (0900 to 2059 h) for treatments 1 to 4, block 2, p l o t t e d on l o g - p r o b a b i l i t y paper . 118 CUMULATIVE PROBABILITY 119 Figure 24: Cumulative distributions of ozone concentrations (ppb) (0900 to 2059 h) for treatments 1 to 4, block 3, plotted on log-probability paper. 120 CUMULATIVE PROBABIL ITY 121 c r i t e r i o n was used (Appendix 4, Table A4-9). In a l l cases high values for p o s i t i v e skewness and p o s i t i v e kurtosis (Appendix 4 f Tables A4-1, A4-4 and A4-7) are caused by a few peak concentrations. It i s c l e a r from the parameters presented i n Appendix 4 that the observations could not have come from a normal d i s t r i b u t i o n . The large p o s i t i v e c o e f f i c i e n t of skewness indicates that the d i s t r i b u t i o n s are skewed with the median value l e s s than the grand mean. These analyses (Appendix 4) also show that the time period during the day i n which O3 concentrations are analysed has a marked influence on the d i s t r i b u t i o n obtained. In turn, t h i s dependence upon the d a i l y time span w i l l influence the s e l e c t i o n of appropriate exposure s t a t i s t i c s , as has been pointed out by Georgopoulos and S e i n f e l d (1982) and Buttazzoni et a l . (1986). 6.2 EXPOSURE STATISTICS The season-long means, peak means and values for other derived exposure s t a t i s t i c s are l i s t e d i n Appendix 3. The d e f i n i t i o n s are presented i n Table 16 (p. 91). The i n t e r - r e l a t i o n s h i p s between the seasonal mean exposure s t a t i s t i c s (Ml[7], Ml[12], M7, M12) and peak exposure s t a t i s t i c s ( P l [ 7 ] , Pl[12], P7, P12) were compared by c a l c u l a t i n g the v a r i a t i o n i n the r a t i o s of the Ml, P l and P7 to the M7 (or comparable M12) f o r each treatment. These were computed separately for both potato and pea as they encompass d i f f e r e n t periods of time. The M7 and M12 were used as the basis of a l l comparisons as they contain the greatest numbers of i n d i v i d u a l ozone measurements. If the a l t e r n a t i v e expressions for exposure are c l o s e l y r e l a t e d to the M7 or 122 M12 res p e c t i v e l y , then the r a t i o s of the s t a t i s t i c s should not vary with the M7 or M12. As i l l u s t r a t e d i n Figures 25 and 26, the seasonal mean r a t i o s , Ml[7]/M7 and Ml[12]/M12 across a l l ozone treatments are quite uniform, whereas the r a t i o s of Pl[7]/M7, P1[12]/M12 and P7/M7 show l i t t l e r e l a t i o n s h i p to the M7 or M12 when computed for either crop (Figures 25 and 26), although the P12/M12 r a t i o s are less v a r i a b l e (Figure 26). 6.3 REGRESSION ANALYSES 6.3.1 PEA 6.3.1.1 Simple l i n e a r regression analyses In s p i t e of the drawbacks with the use of seasonal means as appropriate exposure s t a t i s t i c s , t h e i r widespread use required the computation of simple l i n e a r regressions based upon the M7 and M12 exposure s t a t i s t i c s f o r comparative purposes with the l i t e r a t u r e . The regressions for fresh weights of f i l l e d pods and peas based upon treatment means, are presented i n Figures 27 through 30; the regression c o e f f i c i e n t s are i n Tables 18 and 19. Regardless of the exposure s t a t i s t i c s used (M7 versus M12), the regressions accounted for only 52% of the v a r i a t i o n i n pod weight and only approximately 35% of the v a r i a t i o n i n pea weight. In each case (M7 and M12) an increase i n ozone concentration r e s u l t e d i n a decrease i n y i e l d ; both models were highly s i g n i f i c a n t (Tables 18 and 19). However, neither of these cu r r e n t l y widely used exposure s t a t i s t i c s provided the best f i t t i n g simple l i n e a r functions. A l l models were s i g n i f i c a n t f o r pea and pod y i e l d except for those regressions u t i l i z i n g the sums of concentrations when a threshold was exceeded (SUMxx), the f r a c t i o n a l sums of concentrations above a threshold 123 Figure 25: Relationships between the seasonal mean exposure s t a t i s t i c , M7 and Ml[7], P7 and P l [ 7 ] . Data are expressed as r a t i o s of Ml[7], Pl[7] and P7 to M7. The r a t i o s are computed for J u l i a n days 153 to 211 for pea and 153 to 228/229 for potato. RATIOS OF EXPOSURE STATISTICS ho tx J > cn CT> • x r> 2 RATIOS OF EXPOSURE STATISTICS O KJ J> CT) CO 125 Figure 26: Relationships between the seasonal mean exposure s t a t i s t i c , M12 and Ml[12], P12 and P1[12]. Data are expressed as r a t i o s of Ml[12], P l [ l 2 ] and P12 to M12. The r a t i o s are computed for J u l i a n days 153 to 211 for pea and 153 to 228/229 f o r potato. PEA 12-i CO '~l 1 1 1 1 10 20 30 40 50 M12 OZONE CONCENTRATION [ppb] 127 Figure 27: Relationship between pod fresh weight (g) of peas cv. Puget and ozone expressed as M7 (ppb). The p r e d i c t i o n l i n e was ca l c u l a t e d from the regression equation Y = a + b(M7). = 95% confidence l i m i t s . T 128 129 Figure 28: Relationship between pea fresh weight (g) of peas cv. Puget and ozone expressed as M7 (ppb). The p r e d i c t i o n l i n e was calculated from the regression equation Y = a + b(M7). = 95% confidence l i m i t s . 130 131 Figure 29: Relationship between pod f r e s h weight (g) of peas cv. Puget and ozone expressed as M12 (ppb). The p r e d i c t i o n l i n e was calculated from the regression equation Y = a + b(M12). = 95% confidence l i m i t s . 133 Figure 30: Relationship between pea fresh weight (g) of peas cv. Puget and ozone expressed as M12 (ppb). The p r e d i c t i o n l i n e was calculated from the regression equation Y = a + b(M12). = 95% confidence l i m i t s . I I I 10 20 30 40 M12 OZONE CONCENTRATION [ppb] Table 18. Simple l i n e a r regressions of pod y i e l d s on various exposure s t a t i s t i c s . ( Y i e l d = a + b[Exposure]) Exposure a b r 2 RMS s t a t i s t i c (g) D25 2 (153--179)** 120.2045 -3.0158 0.901 41.0404 0.000 D25 1 (153--211) 133.6962 -1.7426 0.897 43.0055 0.000 2C25 107.8126 -1.3672 0.802 82.3171 0.000 3C25 103.7227 -1.3705 0.778 92.2363 0.000 4C25 96.6258 -1.3257 0.730 112.0645 0.000 D25 4 (203--211) 99.9502 -12.6393 0.680 133.3138 0.000 D25 3 (180--202) 88.0100 -2.4393 0.591 170.0322 0.001 SUMALL 113.1486 -2.0201 0.587 171.6218 0.001 GH12 110.7208 -2.6923 0.579 174.8875 0.002 H12 103.5403 -1.8769 0.553 185.8939 0.002 GM12 108.1286 -2.4190 0.540 191.1790 0.003 H10 99.8128 -2.1372 0.523 198.3866 0.004 M12* 102.5479 -1.6156 0.519 199.8156 0.004 Ml [7] 100.6791 -1.2063 0.516 201.0301 0.004 M7* 102.5145 -1.8465 0.515 201.4577 0.004 GH10 101.9834 -2.8346 0.511 203.3511 0.004 D50 78.9563 -0.9808 0.500 207.7994 0.005 M1[12] 94.8609 -0.7834 0.491 211.7901 0.005 H14 100.0835 -1.4866 0.476 218.0060 0.006 GH14 104.0383 -2.0856 0.460 224.5762 0.008 Pl[7] 93.3868 -0.3624 0.450 228.7824 0.009 P12 98.0995 -0.6878 0.379 258.1079 0.019 P7 96.1818 -0.7224 0.345 272.2519 0.027 IE80 68.6357 -0.0049 0.273 302.3713 0.055 SUM80 69.0551 -0.0160 0.270 303.4191 0.057 IE50 68.3729 -0.0019 0.262 306.6830 0.061 D80 67.1071 -1.2094 0.250 312.0139 0.069 HGT80 67.6728 -0.5207 0.250 310.0301 0.066 SUM100 67.6401 -0.0207 0.233 318.8540 0.080 Pl[12] 74.4678 -0.0945 0.232 319.1886 0.081 IE100 67.6437 -0.0070 0.231 319.8359 0.082 HGT100 66.7033 -0.9596 0.208 329.2129 0.101 SUM50 65.9402 -0.0055 0.198 333.5407 0.111 Whole season; the numbers i n parentheses are J u l i a n days. 2Vegetative growth period. ^Flowering, pod-set. 4 P o d - f i l l i n g . *exposure s t a t i s t i c s used for regressions depicted i n Figures 27 and 29. **exposure s t a t i s t i c s used for regression depicted i n Figure 31. Table 19. Simple linear regressions of pea yields on various exposure statistics. (Yield = a + b[Exposure]) Exposure a b r 2 RMS p statistic (g) D252 (153--179)** 45.8323 -1.1204 0.807 12.3711 0.000 D251 (153--211) 50.8350 -0.6472 0.802 12.6763 0.000 2C25 40.7876 -0.4959 0.685 20.2215 0.000 3C25 39.3129 -0.4974 0.665 21.4751 0.000 4C25 36.2036 -0.4630 0.578 27.0382 0.002 D254 (203--211) 37.3016 -4.3957 0.533 29.9230 0.003 D253 (180--202) 33.3989 -0.8683 0.486 32.9474 0.006 GH12 40.2582 -0.8964 0.417 37.3837 0.013 SUMALL 40.9493 -0.6683 0.417 37.3633 0.013 H12 37.4210 -0.6067 0.375 40.0670 0.020 M1[12] 35.2459 -0.2664 0.368 40.4886 0.021 Ml[7] 36.8252 -0.3991 0.367 40.5777 0.022 GM12 38.8157 -0.7776 0.362 40.8723 0.023 M12* 37.1700 -0.5247 0.355 41.3049 0.024 D50 29.6276 -0.3241 0.354 41.3835 0.025 H10 36.0901 -0.6845 0.348 41.7889 0.026 M7* 36.9668 -0.5919 0.344 42.0691 0.028 GH10 36.6042 -0.8963 0.331 42.8486 0.031 Pl[7] 34.3123 -0.1189 0.314 43.9597 0.037 P7 36.8872 -0.2680 0.308 44.3376 0.039 H14 35.8267 -0.4639 0.300 44.8295 0.042 P12 36.5279 -0.2370 0.292 45.3670 0.046 GH14 36.8993 -0.6435 0.284 45.8812 0.050 SUM80 26.8833 -0.0060 0.248 48.1970 0.070 SUM100 26.4417 -0.0080 0.224 49.7206 0.087 Pl[12] 28.6883 -0.0343 0.198 51.3977 0.111 IE80 26.2029 -0.0016 0.192 51.8072 0.118 IE100 26.0272 -0.0024 0.177 52.7375 0.134 HGT80 25.6802 -0.1608 0.157 54.0029 0.160 D80 25.4499 -0.3665 0.149 54.5601 0.174 IE 50 25.7421 -0.0006 0.148 54.6195 0.175 HGT100 25.5083 -0.3097 0.141 55.0740 0.187 SUM50 24.9749 -0.0016 0.107 57.2381 0.254 Whole season; the numbers in parentheses are Julian days. 2Vegetative growing period. ^Flowering, pod-set. *Pod-filling. *exposure statistics use for regressions depicted in Figures 28 and 30. **exposure statistics used for regression depicted in Figure 32. 137 (IExx) and the sums of hours above a threshold (HGTxx). A l l the s t a t i s t i c s r e l a t e d to i n d i v i d u a l peaks ( P l [ 7 ] r Pl[12], P7 and P12]) ranked low i n terms of r 2 values; the Pl[12] regressions were not s i g n i f i c a n t f or either y i e l d v a r i a b l e . The regressions based on the numbers of days on which the concentration exceed 80 ppb (D80) were not s i g n i f i c a n t for either pea or pod y i e l d . C l e a r l y , as demonstrated by the D25 exposure s t a t i s t i c and i t s variants, for t h i s crop, i t i s the frequency of episodes above what i s considered normal for ambient a i r (25 ppb), and the duration of these episodes, that i s important i n determining the plants response to 0 3, rather than peak concentration per se. Simple l i n e a r regression l i n e s of pod y i e l d s and pea y i e l d s based on the b e s t - f i t t i n g exposure s t a t i s t i c (D25 2; 153-179 days) are presented i n Figures 31 and 32. The regressions account for 90% of the v a r i a t i o n i n pod weight and 81% of the v a r i a t i o n i n pea weight. The comparable D25 s t a t i s t i c for the whole season (D25 1) performed almost equally well (Tables 18 and 19), and somewhat better than those computed for the pod-setting and p o d - f i l l i n g stages. 6.3.1.2 Polynomial regression analyses In these regressions, 21 of the 30 exposure s t a t i s t i c s had s i g n i f i c a n t parameters for the quadratic term. As with the simple l i n e a r regressions, the s t a t i s t i c s based upon concentrations above a threshold performed poorly i n quadratic models, although for a l l exposure s t a t i s t i c s computed there was substantial improvement i n f i t with the a d d i t i o n of the quadratic term for both pod and pea y i e l d s (Tables 20 and 21). 138 Figure 31: Relationship between pod fresh weight (g) of peas cv. Puget and ozone exposure expressed as D25 2 (calculated over J u l i a n days 153-179) (ppb). The p r e d i c t i o n l i n e was calculated from the regression equation: Y = a + b(D25 2). - 9 5 % confidence l i m i t s . 140 Figure 32: Relationship between pea fresh weight (g) of peas cv. Puget and ozone exposure expressed as D25 2 (calculated over J u l i a n days 153-179) (ppb). The p r e d i c t i o n l i n e was ca l c u l a t e d from the regression equation: Y = a + b(D25 2). = 95% confidence l i m i t s . PEA FRESH WEIGHT / PLANT g Table 20. Polynomial regressions of pod y i e l d s on various exposure s t a t i s t i c s . ( Y i e l d = a + ^[exposure] +b 2[exposure] 2) Exposure a b-|_ b 2 r 2 RMS s t a t i s t i c (g) D25 2 (153--179)*136.2947 -5 .6306 D25 1 (153-•211) 168.7823 -4 .0105 GH12 236.3908 -16 .2745 4C25 121.9279 -4 .0594 H12 219.8903 -12 .3695 SUMALL 254.7143 -13 .0974 2C25 131.8744 -3 .4835 GM12 249.7659 -16 .7383 3C25 126.9840 -3 .5989 Ml [7] 212.7082 -8 .1201 M12 222.7882 -10 .9728 M7 224.6660 -12 .6208 H14 232.1353 -11 .3515 H10 198.7970 -12 .6716 D25 4 (203--211) 134.0565 -41 .7650 M1[12] 171.3021 -4 .4067 GH10 212.5264 -17 .2785 GH14 256.1277 -16 .2387 D50 95.9647 -3 .6007 P12 215.5563 -5 .0676 Pl[73 163.4850 -1 .9550 D25 J (180--202) 98.7708 -5 .7630 HGT80 79.5578 -3 .0089 P1[12] 105.6218 -0 .4657 P7 94.1217 -0 .4722 IE80 77.7044 -0 .0216 HGT100 75.5498 -3 .7220 D80 72.9105 -4 .8349 IE100 74.7636 -0 .0243 SUM50 73.1467 -0 .0222 SUM80 74.2602 -0 .0484 SUM100 71.9388 -0 .0613 0 .0819 0. 932 30 .856 0. 000 0 .0318 0. 928 32 .537 0. 000 0 .3407 0. 908 41.602 0. 000 0 .0543 0. 908 41 .833 0. 000 0 .2145 0. 905 43 .196 0. 000 0 .2019 0. 895 47 .516 0. 000 0 .0357 0. 888 50 .842 0. 000 0 .3356 0. 886 51 .535 0. 000 0 .0404 0. 886 51.614 0. 000 0 .0962 0. 882 53.363 0. 000 0 .1658 0. 882 53.606 0. 000 0 .2168 0. 874 57 .093 0. 000 0 .1678 0. 857 64 .737 0. 000 0 .2532 0. 842 71 .600 0. 000 4 .9801 0. 842 71 .643 0. 000 0 .0376 0. 819 82 .117 0. 000 0 .4318 0. 817 82 .934 0. 000 0 .3057 0. 798 91 .604 0. 000 0 .0607 0. 739 118 .403 0. 001 0 .0374 0.707 132 .806 0. 001 0 .0079 0.700 149 .812 0. 002 0.1607 0.686 142 .291 0. 002 0 .0513 0. 508 223 .105 0. 020 0 .0009 0. 475 238.299 0. 029 -0 .0010 0. 453 248 .112 0. 036 0 .0000 0. 441 253 .435 0. 041 0 .1022 0. 392 275 .764 0. 065 0 .1855 0. 373 284 .497 0. 077 0 .0000 0.367 287 .306 0.081 0 .0000 0. 354 292 .822 0. 090 0 .0000 0. 343 297.942 0. 099 0 .0000 0.288 .322 .886 0.154 •"•Whole season; numbers i n parentheses are J u l i a n days. 2Vegetative growing period. ^Flowering, pod-set. 4 P o d - f i l l i n g . *exposure s t a t i s t i c used f o r regression depicted i n Figure 32. Table 21. Polynomial regressions of pea y i e l d s on various exposure s t a t i s t i c s . ( Y i e l d = a + ^[exposure] +b 2[exposure] 2 Exposure a b 1 b 2 r 2 RMS p s t a t i s t i c (g) D25 1 (153--211)*78.9404 -2.4638 D25 2 (153--179) 58.5962 -3.1946 4C25 49.7269 -1.9241 3C25 52.8983 -1.7988 2C25 55.7869 -1.8151 M12 92.6544 -4.8426 Ml [7] 87.5266 -3.5281 SUMALL 107.0855 -5.8434 M7 93.2542 -5.5566 H12 89.2943 -5.2847 GM12 102.8679 -7.2533 GH12 94.7940 -6.7905 H14 95.5043 -4.9221 M1[12] 68.9865 -1.8657 D25 4 (203--211) 53.7275 -18.4231 H10 80.9185 -5.4554 P12 91.8704 -2.3006 GH14 107.5353 -7.2168 GH10 85.1390 -7.2380 D50 37.7418 -1.5739 D25 3 (180--202) 39.1674 -2.6504 Pl[7] 67.9737 -0.8837 HGT80 30.604 -1.1916 P1[12] 39.6208 -0.1645 IE80 30.0287 -0.0086 HGT100 29.3591 -1.5121 IE100 29.2250 -0.0102 P7 36.3921 -0.2078 SUM80 28.6038 -0.0167 SUM50 28.0198 -0.0086 D80 27.8299 -1.8533 SUM100 27.8373 -0.0211 0. 0254 0. 934 4 .603 0. 000 0. 0649 0. 932 4 .736 0. 000 0. 0290 0. 907 6 .524 0. 000 0. 0236 0. 904 6 .711 0. 000 0. 0223 0. 901 6 .914 0. 000 0. 0765 0. 856 10 .059 0. 000 0. 0435 0. 853 10 .278 0. 000 0. 0943 0. 853 10 .268 0. 000 0. 0999 0. 838 11.351 0. 000 0. 0956 0. 829 11 .986 0. 000 0. 1518 0. 822 12 .475 0. 000 0. 1478 0. 819 12.688 0. 000 0. 0758 0. 806 13 .554 0. 000 0. 0166 0. 783 15 .154 0. 000 2. 3985 0. 778 15 .528 0. 000 0. 1147 0. 773 15 .884 0. 000 0. 0176 0. 765 16 .465 0. 000 0. 1420 0. 757 16 .966 0. 000 0. 1896 0. 714 19 .967 0. 001 0. 0290 0.707 20 .500 0. 001 0. 0861 0. 663 23 .529 0. 003 0. 0038 0. 643 24 .950 0. 004 0. 0212 0. 440 39 .156 0. 041 0. 0000 0. 392 42 .536 0. 065 0. 0000 0. 386 42 .916 0.068 0.0445 0. 367 44.283 0. 081 0. 0000 0. 355 45 .104 0. 090 -0. 0002 0.349 45 .544 0. 095 0. 0000 0. 300 48 .967 0. 141 0. 0000 0.288 49 .758 0.154 0. 0761 0. 283 50 .122 0. 160 0. 0000 0. 262 51.610 0. 188 •••Whole season; number i n parentheses are J u l i a n days. 2Vegetative growing period. 3Flowering, pod-set. 4 P o d - f i l l i n g . •exposure s t a t i s t i c used f or regression.depicted i n Figure 33. 144 The D25 2 s t a t i s t i c again performed best. L i t t l e improvement was found for pod y i e l d s when a quadratic model was tested with t h i s s t a t i s t i c (Table 20). However for pea y i e l d s , there was a 10% improvement i n the c o e f f i c i e n t of determination f o r the D25 2 (153-179) and D25 1 (153-211) s t a t i s t i c s (Table 21). The best f i t t i n g quadratic regressions are depicted i n Figures 33 and 34. 6.3.1.3 Stepwise multiple l i n e a r regression analyses Even though the seasonal mean exposure s t a t i s t i c s provided reasonable f i t s when regressed with the pea y i e l d data they do not lead to a s a t i s f a c t o r y understanding of the r e l a t i o n s h i p between concentration frequency and duration of exposure and plant response. To inve s t i g a t e these various components of pollutant exposure further, stepwise multiple l i n e a r regressions were computed using the following independent v a r i a b l e s : D25 1: number of days on which a 1-h mean > 25 ppb (0900-2059 h) occurred; M12I: the i n t e g r a l of the 1-h means (0900-2059 h) for the season; P1[12]: the maximum 1-h mean (0900-2059 h) during the-season; 2C25: number of days with 2 consecutive hourly means > 25 ppb; 3C25: number of days with 3 consecutive hourly means > 25 ppb; 4C25: number of days with 4 consecutive hourly means > 25 ppb. 145 Figure 33: Relationship between pea cv. Puget, pod fresh weight (g) and ozone exposure expressed as D25 2 (calculated over J u l i a n days 153-179). The p r e d i c t i o n l i n e was c a l c u l a t e d from the regression equation: Y = a + bi_-(D252) + b 2 ( D 2 5 2 ) 2 . = 95% confidence l i m i t s . 147 Figure 34: Relationship between pea cv. Puget, pea f r e s h weight (g) and ozone exposure expressed as D25 1 (calculated over J u l i a n days 153-211). The p r e d i c t i o n l i n e was calculated from the regression equation: Y = a + b 1(D25 1) + b 2 ( D 2 5 1 ) 2 . = 95% confidence l i m i t s . PEA FRESH WEIGHT / PLANT g 149 The f i r s t 3 independent v a r i a t e s are analogous to those used i n Nosal's (1983) mu l t i v a r i a t e polynomial Fourier exposure-response model. Only the following two models were selected as being s i g n i f i c a n t : Pod fresh weight (g): Y i e l d = 163.0 - 4.20(D25) + 3.30(2C25) - 1.28(3C25); ( r 2 = 0.95; p =0.000) Pea fresh weight (g): Y i e l d ~ 67.9 - 1.85(D25) + 1.50(2C25) - 1.17(3C25) + 0.67(4C25); ( r 2 = 0.95; p = 0.000). In each case, the number of days on which the threshold of 25 ppb was exceeded (D25 1) contributed most to the r 2 value, accounting for 90% of the t o t a l pod weight v a r i a t i o n and for 80% of the pea weight v a r i a t i o n . Neither the integrated exposure term (M12I) nor the 1-h peak (PI[12]) contributed s i g n i f i c a n t l y i n either case, but several of the sequential hour terms were included. It should be noted that the c o e f f i c i e n t s f o r the 2C25 and 4C25 terms are p o s i t i v e . 6.3.2 POTATOES 6.3.2.1 Simple l i n e a r regression analyses Simple l i n e a r regressions for f r e s h weights of tubers, based upon the M7 and M12 seasonal means, are presented i n Figures 35 and 36; the regression c o e f f i c i e n t s are i n Table 22. The regressions accounted for 52% of the v a r i a t i o n i n tuber weight, regardless of which exposure s t a t i s t i c was used. However, as was the case for pea y i e l d s , the use of these seasonal means d i d not r e s u l t i n the best f i t t i n g simple l i n e a r functions (Table 22). As with pods and peas, few of the models f i t t e d using sums of 150 Figure 35: Relationship between potato cv. Russet Burbank tuber fresh weight (g) and ozone exposure expressed as M7 (ppb). The p r e d i c t i o n l i n e was c a l c u l a t e d from the regression equation: Y = a + b(M7). = 95% confidence l i m i t s . . 151 152 Figure 36: Relationship between potato cv. Russet Burbank tuber fresh weight (g) and ozone exposure expressed as M12 (ppb). The p r e d i c t i o n l i n e was calculated from the regression equation: Y = a + b(M12). = 95% confidence l i m i t s . 154 Table 22. Simple linear regressions of potato tuber yields on various exposure statistics. (Yield = a + b[Exposure]) Exposure a b r 2 RMS statistic (g) GH12** D254 (212-228,229) HR12 D251 (153-228,229) GM12 Ml [7 ] M7* M12* SUMALL M1[12] 2C25 H14 H10 Pl[7] GH10 GH14 4C25 3C25 D50 D253 D252 HGT80 D80 IE50 P1[12] IE80 SUM50 P12 P7 HGT100 SUM80 IE100 SUM100 (178-211) (153-177) 1239 1040 1201 1285 1222 1175 1183 1191 1257 1138 1113 1183 1154 1116 1167 1205 1039 1082 958 1057 1152 879 874 885 956 877 868 1097 1075 859 866 858 836 .201 .287 .432 .706 .183 .614 .356 .808 .380 .555 .224 .311 .258 .924 .360 .650 .180 .949 .416 .496 .034 .329 .895 .968 .157 .865 .247 .597 .173 .465 .522 .220 .804 -23.3833 -24.5595 -17.5688 -8.8892 -20.9014 -11.3950 -16.8840 -15.2183 -13.8809 -7.9591 -7.1798 -14.1230 -19.3772 -3.4234 -24.4534 -18.6500 -7.1339 -7.2628 -7.1243 -16.2878 -20.0248 -4.9796 -11.6403 -0.0172 -1.0361 -0.0416 -0.0547 -5.4892 -5.6153 -8.3377 -0.1132 -0.0559 -0.1319 0. 0. 0. 0. 0.568 0.557 ,537 .532 .529 ,524 0.521 0.520 0.518 0.514 0.501 0.499 0.495 0.491 0.491 0.491 0.484 0.482 0.473 0.433 0.421 0.349 0.343 0.343 0.318 0.310 0.308 0.304 0.258 0.231 0.211 0.203 0.114 15675.06 16096.69 16815.67 16996.17 17082.57 17299.32 17400.19 17438.31 17491.34 17661.86 18112.89 18180.58 18342.40 18489.84 18475.97 18471.63 18741.54 18804.21 19124.72 20579.52 21034.15 23621.53 23859.84 23857.97 24747.13 25063.46 25128.87 25268.19 26953.97 27921.06 28657.12 28933.87 32162.55 0.002 0.002 0.003 0.003 0.003 003 004 004 004 004 005 005 005 005 005 005 006 006 0.007 0.011 0.012 0.026 0.028 0.028 0.036 0.039 0.040 0.041 0.064 0.082 0.099 0.106 0.238 Whole season; number in parentheses are Julian days. 2Vegetative growing period. ^Flowering, and tuber initiation. 4Tuber f i l l i n g . •exposure statistics used for regressions depicted in Figures 35 and 36. **exposure statistics used for regression depicted in Figure 37. 155 concentrations, hours above a threshold or f r a c t i o n a l sums above a threshold were s i g n i f i c a n t . Peak exposure s t a t i s t i c s a lso ranked low i n terms of importance, with r 2 values mostly l e s s than 0.50. The exposure s t a t i s t i c that performed best o v e r a l l was GH12 ( r 2 = 0.57, Figure 37), the seasonal geometric mean of the d a i l y means occurring between 1200 and 1259 h. Of the D25 s t a t i s t i c s , the one which performed best for tuber y i e l d (lowest RMS and highest r 2 ) was that computed over the time period where tuber f i l l i n g was taking place (D25 4). This contrasts with the s i t u a t i o n found for peas, where exposure during the vegetative period of growth was important. 6.3.2.2 Polynomial regression analyses Only the season-long geometric hourly mean (GH10) computed for the hour between 1000-1059 h, resulted i n both parameters being s i g n i f i c a n t i n the f i t t e d quadratic models for potato tuber y i e l d . Only moderate improvement resu l t e d from the add i t i o n of a quadratic term for the other models examined (Table 23). In a number of cases (P12, P1[12], D80 e t c . ) , the introduction of a quadratic term res u l t e d i n non-s i g n i f i c a n t models (compare Tables 22 and 23). 6.3.2.3 Stepwise multiple l i n e a r regression analysis Stepwise multiple l i n e a r regression, using the same independent va r i a b l e s as for pea y i e l d s , rejected a l l models except the simple l i n e a r regression: Y i e l d = 1285.7 - 8.89(D25 1); already noted i n Table 22. 156 Figure 37: Relationship between potato cv. Russet Burbank tuber fresh weight (g) and ozone exposure expressed as GH12 (ppb). The p r e d i c t i o n l i n e was calculated from the regression equation: Y = a + b(GH12). = 95% confidence l i m i t s . 157 158 Table 23. Polynomial regressions of potato tuber yields on various exposure statistics. (Yield = a + b^[exposure] +b2[exposure]2) Exposure a b 1 b 2 r 2 RMS statistic (g) GH10 1827.4109 -111.565 2.6087 0.650 13845.57 0.003 H10 1744.1877 -84.168 1.6000 0.629 14682.05 0.004 GM12 1871.4728 -85.905 1.4968 0.623 14918.77 0.005 GH12 1694.2165 -73.662 1.2734 0.622 14978.51 0.005 M7 1740.9781 -66.687 1.0083 0.610 15438.97 0.006 GH14 1988.8393 -89.316 1.4714 0.607 15563.67 0.006 D254(212-228,229) 1139.5612 -55.236 1.5908 0.604 15698.26 0.006 D251(153-228,229) 833.9766 13.182-0.2318 0.586 16391.42 0.008 M12 1684.8305 -54.290 0.7039 0.585 16451.01 0.008 H14 1767.5910 -56.984 0.7171 0.585 16421.66 0.008 SUMALL 1884.6081 -52.425 0.5516 0.580 16644.60 0.009 H12 1589.0857 -57.762 0.7254 0.580 16625.83 0.008 Ml[7] 1583.1711 -36.833 0.3576 0.578 16726.62 0.009 D50 1044.4897 -17.778 0.1977 0.546 17965.46 0.013 M1[12] 1323.4030 -17.107 0.0996 0.533 18478.34 0.015 4C25 1116.3515 -13.280 0.0900 0.502 19728.80 0.022 2C25 1088.8186 -5.564 -0.0204 0.502 19714.54 0.022 Pl[7] 1227.1213 -5.924 0.0124 0.496 19949.29 0.023 3C25 1093.8256 -8.037 0.0105 0.482 20502.60 0.027 P7 1027.4203 -1.736 -0.0141 0.481 20536.67 0.027 D252(153-177) 974.6010 13.541 -1.1674 0.477 20711.12 0.028 D253(178-211) 1055.1589 -15.855 -0.0147 0.433 22449.64 0.044 HGT80 925.8212 -13.908 0.1624 0.394 23986.19 0.063 SUM50 907.5500 -0.134 0.0000 0.364 25182.51 0.083 D80 895.9469 -23.924 0.5604 0.363 25225.21 0.084 P1[12] 1043.7781 -2.077 0.0024 0.340 26135.77 0.102 P12 1395.1237 -16.373 0.0908 0.330 26530.50 0.110 IE80 903.3038 -0.083 0.0000 0.325 26735.34 0.115 HGT100 899.9969 -20.052 0.3896 0.278 28589.65 0.167 SUM80 880.4245 -0.184 0.0000 0.216 31061.97 0.263 IE100 877.6498 -0.100 0.0000 0.215 31084.99 0.264 SUM100 850.8229 -0.265 0.0001 0.121 34822.77 0.493 Whole season; numbers in parentheses are Julian days. 'Vegetative growing period 'Flowering and tuber initiation Tuber f i l l i n g 159 7. DISCUSSION 7.1 Background The Lower Mainland of B r i t i s h Columbia and surrounding r u r a l areas are recognized as one of the regions i n Canada where exceedance of the Canadian Maximum A i r Quality Objectives i s not uncommon (Wilson et a l . , 1984). This area represents important and diverse a g r i c u l t u r a l lands i n B.C., yet no information e x i s t s regarding the e f f e c t s of O3 on crops grown there. It i s known that O3 a l t e r s the growth and y i e l d of p l a n t s . However, the e f f e c t s and magnitude of response have been shown to vary with species and c u l t i v a r , environmental conditions, and c u l t u r a l conditions. U n t i l recently, most of the long-term studies on the y i e l d e f f e c t s of O3 have exposed plants to constant rather than f l u c t u a t i n g concentrations representative of f i e l d conditions (Roberts, 1984). Although the large-scale studies conducted i n the National Crop Loss Assessment Network (NCLAN) program have been aimed at obtaining ozone response functions for a range of important crops; the methodology used (open-top chambers; constant or proportional additions of ozone to ambient a i r , etc.) has f a i l e d to produce experimental conditions t r u l y resembling those l i k e l y to occur i n the f i e l d . For example, using t h e i r exposure protocols those treatments r e c e i v i n g high exposures always receive high concentrations regardless of whether or not the additions are constant or proportional. In addition, due to the presence of the chambers, plants are constantly being subjected to the a i r movements needed to provide adequate mixing of the a i r within the chambers and to maximise mass flow through the chambers. The plants within the chambers thus never experience the v a r i a t i o n s i n wind speed that t y p i f y true f i e l d conditions. 160 The ZAPS approach was u t i l i z e d i n the present experiments to provide r e a l i s t i c enrichment of the a i r surrounding the crop without the microclimatic e f f e c t s imposed by chambers. Because of changes i n windspeed and d i r e c t i o n and the proximity of the p l o t s to each other, the plants are subjected to widely varying exposures more c h a r a c t e r i s t i c of those of ambient a i r . 7.2 A i r Quality Since the o v e r a l l i n t e n t i o n i n these experiments was to simulate what might be happening i n an ambient a i r s i t u a t i o n , i t was important to determine the nature of the exposures achieved, both temporally and s p a t i a l l y , and to ascertain that the d i s t r i b u t i o n s of ozone concentrations observed simulated those l i k e l y to occur at elevated ambient O3 l e v e l s . V e r t i c a l p r o f i l e s of O3 concentrations were measured i n the absence of a plant canopy i n both 1985 and 1986. As was an t i c i p a t e d , 0 3 was dispersed down to 10 cm (the lowest point measured). The v a r i a t i o n s observed around the mean 0 3 concentrations (Figures 3 and 13) are to be expected as t h i s i s an open-air system. Factors contributing to t h i s v a r i a t i o n include changes i n the p r e v a i l i n g wind speed and d i r e c t i o n , a i r temperature structure, surface f r i c t i o n and absorption (Runeckles et a l . , 1981a). The decrease i n mean concentration at the 80 and 90 cm l e v e l i n 1985 and 1986 r e s p e c t i v e l y may be due to v a r i a t i o n i n windspeed and interference with mixing caused by the manifolds themselves, and the l o c a t i o n of the sampling l i n e i n l e t r e l a t i v e to the nearest discharge o r i f i c e s . A decrease i n p o l l u t a n t concentration near the s o i l surface (which acts as a sink f or 0 3 , Turner et a l . 1973, 1974) and s l i g h t 161 increases or decreases i n concentration at or just below the manifold were al s o found by Runeckles et a l . (1981a) using the same system. Though the v e r t i c a l p r o f i l e s were not measured i n the presence of a plant canopy, from the observations of Bennett and H i l l (1975) and McLeod et a l . (1985), the concentration would be expected to decrease with decreasing height i n the presence of a plant canopy, as a r e s u l t of f o l i a r uptake and v e n t i l a t i o n of the canopy. The v e r t i c a l p r o f i l e s measured i n the present studies i n d i c a t e that, i n s pite of being discharged from point sources along the arms of the manifolds, the O3 released mixed r a p i d l y with the ambient a i r , with no evidence of l o c a l i z e d high concentrations. Horizontal p r o f i l e s were obtained i n 1986 only. The data presented i n Table 24 suggest that the supplementary 0 3 provided by the ZAPS used i n the present study (with h o r i z o n t a l discharge) was reasonably uniformly d i s t r i b u t e d over the i n d i v i d u a l p l o t s , during the 4-hour periods during which the measurements were made.. In contrast, Lee and Lewis (1977) examined the h o r i z o n t a l d i s t r i b u t i o n of S0 2 produced by t h e i r l arge-scale ZAPS, with o r i f i c e s discharging v e r t i c a l l y downwards, and observed that locations near the d e l i v e r y pipes had concentrations approximately three times higher than locations at the midpoints between pipes. They also found that concentrations were strongly correlated with the inverse of windspeed and increased i n the d i r e c t i o n of the p r e v a i l i n g wind. The combined findings i n the present studies with respect to the v e r t i c a l and h o r i z o n t a l d i s t r i b u t i o n s of O3, therefore, confirm that the ZAPS approach i s capable of modifying the composition of the a i r to which 162 crops are exposed, i n a reasonably predictable, measurable and consistent manner. For both 1985 and 1986 the season-long pattern of the average 0 3 concentrations demonstrated a f a i l u r e to achieve p r o p o r t i o n a l i t y (Figures 6 and 14-16). As outlined i n Section 4.1.2, i n 1985 t h i s was l a r g e l y due to mechanical problems. However, the di u r n a l pattern of O3 concentrations, on a seasonal basis, c l e a r l y resembled that of ambient a i r for both years. The modifications introduced i n 1986 resulted i n considerably improved tracking of the ambient a i r . In 1986, the plot s within blocks were randomly assigned regardless of the p r e v a i l i n g wind d i r e c t i o n and carry-over occurred. As a r e s u l t , the season-long means for the i n d i v i d u a l p l o t s d i d not ne c e s s a r i l y follow the same order as the number of o r i f i c e s i n the manifolds above them suggested. However, such seasonal averages do not ne c e s s a r i l y r e f l e c t the s i t u a t i o n that may be occurring on any one day. To develop exposure-response r e l a t i o n s h i p s that are relevant f o r the sett i n g of ambient a i r q u a l i t y standards, an appropriate summary of ambient a i r O3 d i s t r i b u t i o n s i s required. Many a i r p o l l u t i o n studies are based on the assumption that p o l l u t i o n concentrations are well described by either a normal or log-normal d i s t r i b u t i o n (Larsen, 1976; Male, 1982). However, evidence i s a v a i l a b l e which contradicts the assumptions of normal or l o g -normal d i s t r i b u t i o n s and suggests that Weibull, gamma and more highly parameterized log-normal d i s t r i b u t i o n s such as the Johnson S B may be more appropriate (Larsen 1973; Nosal, 1984; Taylor et a l . , 1986). In the present study, Log-normal, Weibull, gamma and Johnson S B (4-parameter log-normal) d i s t r i b u t i o n s were found to f i t the ambient a i r data, depending on the averaging time and number of exposure hours i n a 24-h 163 period that were used (Appendix 4). Using a complete 24-h data set r e s u l t s i n the i n c l u s i o n of the large number of low concentrations t y p i c a l l y occurring at night. In t h i s case, the 2-min averages appear to be best described by a Weibull d i s t r i b u t i o n (Appendix 4, Table A4-6) while the data condensed to 1-h averages favors the gamma or Johnson S B d i s t r i b u t i o n s (Appendix 4, Table A4-9). On the other hand, the 2-min averages occurring between 0700 and 2059 were best described by a simple log-normal d i s t r i b u t i o n (Appendix 4, Table A4-3). Although the 0 3 data for a l l treatments i n both 1985 and 1986 are not p r e c i s e l y log-normal (Figures 4, 17-20), nevertheless the concentrations f o r both years c l e a r l y have the same general form, i n d i c a t i n g that the ZAPS was re l e a s i n g O3 to provide elevated concentration d i s t r i b u t i o n s resembling those occurring i n ambient a i r . This contrasts with the d i s t r i b u t i o n s found i n some NCLAN studies in v o l v i n g open-top chambers which have been shown to y i e l d a t y p i c a l bimodal d i s t r i b u t i o n s at the higher rates of O3 addition (Heagle et a l . , 1987; Lefohn et a l . , 1988). Such bimodal d i s t r i b u t i o n s have not been reported to occur i n ambient a i r . Since they i n d i c a t e a greater than natural frequency of occurrence of exposure concentrations i n the higher range, and because of the importance of peak concentrations i n determining plant response, the responses observed to exposures with such d i s t r i b u t i o n s w i l l be biased. The d i s t r i b u t i o n s observed for the treatments applied i n both 1985 and 1986 i n the present studies are s i m i l a r , and have the same unimodal form as those observed i n ambient a i r . It should be noted that the 1985 d i s t r i b u t i o n s presented i n Figure 4 are for 1-h means between 0700 and 2059 h, whereas those for 1986 i n Figures 17-20 are more d e t a i l e d and are based 164 upon the i n d i v i d u a l 2-minute averages between 0900 and 2059 h recorded for each l o c a t i o n . 7.3 E f f e c t s of ozone on crops Although previous experience (Runeckles et a l . , 1981b) had shown l i t t l e evidence of heterogeneity among the f i e l d blocks, with regard to s o i l and other fac t o r s , i n 1985 i t was c l e a r that a great deal of v a r i a t i o n existed between plants both within and between treatments. These differences and the observed differences i n s o i l water retention, and the incidence of disease, precluded s t a t i s t i c a l demonstration of di f f e r e n c e s due to O3 for most crop parameters (Tables 3 to 6). In addition, without true r e p l i c a t i o n of treatments, d i f f e r e n t i a t i o n between causes due to O3 and those from a b i o t i c ( i e . drainage) and b i o t i c ( i e . pests and disease) e f f e c t s was not pos s i b l e . Consequently, no sound conclusions can be drawn from the 1985 y i e l d data, i n s p i t e of the s t a t i s t i c a l l y s i g n i f i c a n t responses observed for b r o c c o l i , cv's. Emperor and SGI and bean cv. BBL-GV2 using various exposure s t a t i s t i c s (Tables 7 and 8). There are no reports of f i e l d experiments using b r o c c o l i i n the l i t e r a t u r e . Of the work done on crop plants i n the Brassicaceae none of the crops investigated were grown to maturity. In general, the response to O3 was a negative l i n e a r r e l a t i o n s h i p with a p o s i t i v e or negative quadratic term for some of the v a r i a b l e s measured for cauliflower (Marie and Ormrod, 1986). This i s si m i l a r to the findings i n the present study for mature b r o c c o l i plants (Tables 7 and 11). However, these r e s u l t s have to be viewed i n the context of the observed d i f f e r e n t i a l occurrence of club root (Plasmodiophora brassicae Wor.) which only appeared i n the 03-treated p l o t s . While this^ may have been the chance r e s u l t of inoculum being absent from the control p l o t 165 s o i l , the fa c t that q u a l i t a t i v e l y , the i n f e c t i o n seemed to be greater i n the high 03-treated plants suggests that these plants were more s e n s i t i v e to i n f e c t i o n following exposure to O3. Similar findings have been reported for other crops and pathogens (Manning et a l . , 1969; James et a l . , 1980a; Bisessar, 1982). Heggestad et a l . (1980) investigated the y i e l d of several snap bean c u l t i v a r s over a number of years. They found a 14% decrease i n y i e l d on average for cv. BBL-290 from 1972 to 1974 and a 22% decrease i n y i e l d f o r cv. BBL-274 i n 1976 i n n o n - f i l t e r e d as compared with c h a r c o a l - f i l t e r e d a i r . Trends i n d i c a t i n g decreases i n bean y i e l d with increasing exposure to ozone were found i n the present study (as indicated by the negative l i n e a r regression c o e f f i c i e n t s ) although they were not s t a t i s t i c a l l y s i g n i f i c a n t . Haas (1970) found that the timing and s e v e r i t y of symptoms i n bean were a function of the crop maturity and vi g o r . Though no fi r m conclusion can be drawn from the present study i n l i g h t of the confounded experimental design, bean plant vigor may have been reduced i n the AAX1.5 and AAX2.0 pl o t s due to root rot thus making them more s u s c e p i t i b l e to O3, and po s s i b l y accounting for the spurious c u r v i l i n e a r resposne r e l a t i o n s h i p observed (Table 12). Although the e f f e c t s of O3 on a number of pea parameters have been reported under greenhouse or growth chamber conditions (Ormrod, 1976; Olszyk and T i b b i t t s , 1982; Kobriger and T i b b i t t s , 1985), no studies to date have been reported for f i e l d conditions. Furthermore, since the concentrations used i n these other studies were high (0.1-1.0 ppm) and no attempts were made to quantify the e f f e c t s of 0 3 on marketable y i e l d , comparisons with the present study are not p o s s i b l e . 166 The presence of powdery mildew (Erysiphe polygoni DC.) on peas i n the control p l o t s i n the 1985 experiment led to poorer than expected pod y i e l d i n these p l a n t s . Thus the quadratic response observed (Table 13) may be a spurious r e s u l t . This becomes more convincing i n l i g h t of the negetative intercepts for these regression equations. Excluding the control plants from the analysis would no doubt lead to an improvement i n the r e l a t i o n s h i p . Indeed, r e s u l t s from 1986 suggest that the r e l a t i o n s h i p between O3 concentration i s adequately described by a l i n e a r model (Table 18). The lack of severe i n f e c t i o n of powdery mildew on the O^-treated plants suggests that the pathogen i s perhaps s e n s i t i v e to the a i r p o l l u t a n t . This i s supported by Laurence (1981) who indicated that lesions of o b l i g a t e parasites are usually smaller on plants exposed to O3 when compared with control plants. The e f f e c t s of O3 on potatoes have been investigated by a number of researchers. Mosley et a l . (1978) assessed f o l i a r i n j u r y and tuber y i e l d i n 59 potato c u l t i v a r s exposed to O3 i n an open-air system. They found that there was a wide range of responses among the c u l t i v a r s , with the later-maturing v a r i e t i e s being more r e s i s t a n t . In general, they observed a decline i n y i e l d with an increased O3 i n j u r y index. Bisessar (1982), i n an open-air experiment i n which EDU was used as antioxidant protectant of potato plants, found that tuber weights of Norchip were reduced i n untreated plants, and that tuber weight was i n v e r s e l y r e l a t e d to f o l i a r i n j u r y . He also found that A l t e r n a r i s o l a n i (early b l i g h t ) colonized O3-injured s i t e s more than non 0 3 - i n j u r e d areas of the f o l i a g e . This was also supported by the work of Holley et a l . (1985), suggesting that 0 3 i n j u r y was a factor i n increasing the i n f e c t i o n of potato leaves by A l t e r n a r i a 167 s o l a n i . However, i n the present studies, i n f e c t i o n of potato leaves by A l t e r n a r i a s o l a n i i n 1986 appeared, on the basis of v i s u a l inspection, to be greatest i n the control and block 2 plants, suggesting that there was no p r e f e r e n t i a l attack of ozonated plants. A number of researchers have suggested that i n f e c t i o n by A l t e r n a r i a solani i s r e l a t e d to plant senescence (Harrison et a l . , 1965 and Douglas and Pavek, 1972 as c i t e d i n Barclay et a l . , 1973). S i g n i f i c a n t reduction i n the incidence of early b l i g h t was achieved by subjecting potato plants to conditions of high nitrogen and low phosphorus (Barclay et a l . , 1973), and i t was suggested that t h i s was the r e s u l t of delayed maturity of these plants caused by the high nitrogen f e r t i l i t y . It i s therefore possible that, i n the present study, the more vigorously growing plants i n the c o n t r o l block and block 2 (the drainage-poor block) began to suffer from a nitrogen shortage, which speeded up maturity and thus rendered them more susceptible to i n f e c t i o n . Both peas and potatoes were sprayed with p e s t i c i d e i n 1985 and 1986. No spray damage occurred on potatoes. However i n 1986, spray damage, amounting i n some cases to approximately 10% of the leaf area, occurred on many of the 0 3 - t r e a t e d pea p l a n t s . Plants i n the c o n t r o l p l o t s were least a f f e c t e d and those i n block 2 suffered more than the c o n t r o l but less than those i n blocks 1 and 3. These plants (block 2) a l s o looked larger and hea l t h i e r than those of the l a t t e r two blocks. It i s p o s s i b l e that some i n t e r a c t i o n between 0 3 and the p e s t i c i d e occurred, but, t h i s i s not testable i n the present study. The e f f e c t of the spray damage may have been i n biasing the regressions i n favor of the c o n t r o l plants, so that a steeper r e l a t i o n s h i p than normal, may have been achieved. 168 7.4 Exposure s t a t i s t i c s A l l exposure-response models are dependent on the method of expressing exposure. As reviewed i n Section 2.3, to date there i s no consensus on the appropriate mathematical expression of exposure based s o l e l y on ambient a i r concentrations that should be used i n plant resposne modelling. This i s p a r t l y because of the uncertain r e l a t i o n s h i p between exposure and dose or e f f e c t i v e dose (discussed i n Section 2.2), and p a r t l y because of the lack of agreement as to the features that should be incorporated i n t o the expression used to describe exposure. The development of response functions for the crops i n the present study required the i n v e s t i g a t i o n of a range of indices (exposure s t a t i s t i c s ) by which the exposure to O3 could be characterized. Various methods of c h a r a c t e r i z i n g exposure have been described, e.g. mean concentrations with varying averaging times (Heck et a l . , 1982, 1984a, 1984b; Kress et a l . , 1985; Heagle et a l . , 1987); the number of hours or days with hourly concentrations above a threshold (Nosal, 1984; Ashmore, 1984; Lee et a l . , 1987); integrated exposure above a threshold (Oshima, 1976; Lefohn and Benedict, 1982); integrated hourly O3 concentrations ra i s e d to an exponent (Nouchi and Aoki, 1979; Larsen and Heck, 1984) and a sigmoidal weighting function (Lefohn and Runeckles, 1987). In the present study 30 exposure s t a t i s t i c s were computed for use i n developing exposure-response functions for the various crops studied. Of these s t a t i s t i c s 13 are new and have not been previously described i n the l i t e r a t u r e , v i z . D25, D50, D80, 2C25, 3C25, 4C25, GM12, H10, H12, H14, GH10, GH12 AND GH14 (Table 16), while several a d d i t i o n a l s t a t i s t i c s are variants of s t a t i s t i c s used by others, i . e . Ml[12], P12, Pl[12], SUM50 and IE50. 169 The d i f f e r e n t s t a t i s t i c s may be categorized as arithmetic mean s t a t i s t i c s (M7, M12, Ml[7], Ml[12] and HXX); geometric means (GM12 and GHXX) peak s t a t i s t i c s (P7, P12, P14[7] and P1[12]); cumulative indic e s above some threshold, based on concentration (IEXXX, SUMXXX); or time, (DXXX, XC25, HGTXXX); and a cumulative index with no threshold, based on concentration (SUMALL). Each of these groups of exposure s t a t i s t i c s has i t s drawbacks and merits. For instance, arithmetic means are easy to compute, and therefore t h e i r use i n the development of a i r q u a l i t y standards i s appealing. However, t h e i r use i s being questioned on s t a t i s t i c a l grounds because 0 3 a i r q u a l i t y data are not t y p i f i e d by normal d i s t r i b u t i o n s (Runeckles, 1987). Indeed, i n many studies in c l u d i n g the present ones, a i r q u a l i t y has been shown to be only poorly described even by log-normal d i s t r i b u t i o n s (Figures 5 and 21 to 24). Hence, though the season-long arithmetic means are widely used i n the l i t e r a t u r e (Heck et a l . , 1982, 1984a, 1984b; Kress et a l . , 1985; Cure et a l . , 1986; Heagle et a l . , 1987), such exposure s t a t i s t i c s do not adequately summarize the temporal frequency of occurrence of O3 concentrations. This c r i t i c i s m of the use of simple seasonal means i s further backed up by the work of Heagle et a l . (1986) and Lefohn et a l . (1988) where t h e i r use was shown to be inappropriate for comparing y i e l d responses i n soybean and winter wheat, re s p e c t i v e l y , across seasons. Although the seasonal means were found to be si m i l a r the y i e l d s were d i f f e r e n t for each year and the lengths of the experimental exposure periods were a l s o d i f f e r e n t . The use of means d i d not take t h i s i n t o account. Furthermore, f l u c t u a t i n g versus constant concentrations at s i m i l a r t o t a l exposures have been shown to be important i n determining a plant's response to a i r pollutants (Hogsett et a l . , 1985; Male et a l . , 1983; 170 Musselman et a l . , 1983). An arithmetic mean under these circumstances would t r e a t a l l exposure dynamics and concentrations as being equally e f f e c t i v e i n e l i c i t i n g a response,thus ignoring the existence of peaks. In the experiments of Hogsett et a l . (1985), the season-long 7-h mean (M7) was smaller for the f l u c t u a t i n g than constant O3 treatment and yet the y i e l d reduction was greater. Although the use of the GM12 (season-long geometric mean of d a i l y 12-h geometric means) i s more defensible on mathematical grounds, i t s t i l l suffers.from the drawbacks appl i c a b l e to a l l averages. The HXX and GHXX s t a t i s t i c s s t r i v e for s i m p l i c i t y , and ignore the problem of deciding what period of time i n a day the season-long mean should be c a l c u l a t e d . Instead, these s t a t i s t i c s can be selected to u t i l i z e a one-hour time period i n the day where the plant i s considered to be a c t i v e l y photosynthesizing. Presumably O3 dispensed i n t h i s period w i l l have a greater impact than at some other time. However, these means r e a l l y represent truncated versions of the M7 and M12 s t a t i s t i c s and are therefore subsets of the concept of the M7. The season-long geometric hourly mean, GH12 (computed between 1200-1259-h) performed well for the 1986 potato y i e l d data, suggesting that there i s a r e l a t i o n s h i p between exposure and time of day, a f f e c t i n g plant response. It has been pointed out by Salisbury and Ross (1985) that a plant may be more s e n s i t i v e to toxins during a part of i t s c i r c a d i a n c y c l e . Consequently, the timing of p o l l u t a n t episode i n r e l a t i o n to the c i r c a d i a n c y c l e of the plant may be very important i n determining how e f f e c t i v e an exposure i s i n e l i c i t i n g a response. For instance, Reinert et a l . (1972) found that tomato plants exposed i n the early afternoon as opposed to the morning developed more i n j u r y . 171 Because of the c r i t i c i s m s o u t l i n e d above of the value of season-long means as independent variables i n response regressions, other more meaningful exposure s t a t i s t i c s have been proposed. For example, cumulative i n d i c i e s are s e n s i t i v e to the d i f f e r e n t experimental exposure lengths, as they increase as the exposure period increases. As Runeckles (1988) points out, t h i s i s an important consideration when comparing r e s u l t s over seasons. The importance of t h i s f a c t becomes clear i n the studies by Heagle et al.,(1986) who found a d i s p a r i t y between predicted y i e l d reductions i n soybean for two growing seasons using an integrated exposure index (ppm.h) as the exposure s t a t i s t i c but none using the season-long 7-h mean (M7). This i s to be expected since among other possible factors such as c u l t u r a l techniques and meterological conditions, the duration of the exposure d i f f e r e d for the two years. Though the integrated exposure index (ppm.h) i s s e n s i t i v e to the d i f f e r e n t lengths of experimental exposure periods, the season-long 7-h mean i s not. The mere fac t that comparable response functions were obtained using the seasonal mean s t a t i s t i c , whereas d i s t i n c t l y d i f f e r e n t functions were obtained using the integrated exposure index s t a t i s t i c i s i n s u f f i c i e n t j u s t i f i c a t i o n f o r s e l e c t i n g the seasonal mean as a v a l i d exposure s t a t i s t i c . Since the length of season, or exposure period, i s somewhat a r b i t r a r y , and, at l e a s t i n NCLAN usage, has been defined s o l e l y i n terms of the period over which O3 monitoring data were c o l l e c t e d , regardless of the duration of the actual growing period, the use of such mean values could lead to the absurd extreme i n which exposure f o r , say, a 7-day "season" would be expected to have the same impact as exposure for a 70-day season. However, while cumulative indices have c e r t a i n d e s i r a b l e features f o r use as exposure s t a t i s t i c s , they suffer from one of the c r i t i c i s m s that 172 apply to the use of mean and peak s t a t i s t i c s i . e . they are not phenologically s e n s i t i v e and do not i d e n t i f y when, during a crop's development, the episodes of high concentrations occur. Cumulative indices based upon exceedances of threshold concentrations may further be c r i t i z e d f o r ignoring the lower concentrations that have been shown by some researchers to predispose the plant to subsequent i n j u r y (Runeckles and Rosen, 1974; Johnston and Heagle, 1982). The use of peak exposure s t a t i s t i c s a l s o has t h i s l i m i t a t i o n . Lefohn and Runeckles (1987) have indicated that i n situa t i o n s where repeated high peaks with l i t t l e time between exposures occurs the integrated exposure s t a t i s t i c may serve as a useful dose s t a t i s t i c . For the more common s i t u a t i o n i n which the occurrence of peaks i s stochastic, Lefohn and Runeckles (1987) suggested the use of a sigmoidal or l o g i s t i c weighting function that weights a l l indvidual concentrations p r i o r to summing the weighted values over time, unlike the simple integrated exposure index of Lefohn and Benedict (1982). When put to the test on a l i m i t e d number of data sets, i t has been recently shown by Lefohn et a l . (1988) to be no better but no worse an exposure s t a t i s t i c f o r use i n es t a b l i s h i n g the r e l a t i o n s h i p between exposure to O3 and reductions i n the y i e l d of a g r i c u l t u r a l crops than cumulative i n d i c i e s and season-long means. To f a c i l i t a t e comparisons with the l i t e r a t u r e , the 1-h mean (which forms the basis of many a i r p o l l u t i o n standards) was used as the basis f o r computing the widely used seasonal M7 and M12 means. Exposure s t a t i s t i c s which recognize the importance of peak concentrations as stated above include the P7, P l [ 7 ] , P12, Pl[12], Ml[7] and Ml[l2] (Heck et a l . , 1984a, 1984b; Cure et a l . , 1986). One of the problems with these exposure s t a t i s t i c s i s r e l a t i n g y i e l d to a s t a t i s t i c c a l c u lated over a s p e c i f i c 1-h 173 period versus a season-long mean. Peaks vary i n magnitude and time of occurrence and i t i s therefore d i f f i c u l t to see how any i s o l a t e d peak value (whether a 1-h, 7-h or 12-h maximum) can be used as the basis of response models. In t h i s study poor r e l a t i o n s h i p s were found between the M7 or M12 and the r a t i o s of Pl[7]/M7, Pl[12]/M12, P7/M7 and P12/M12 across a l l 0 3 treatments (Figures 25 and 26). This i s not true of NCLAN studies based upon open-top chamber experiments (Heck et a l . , 1984a, 1984b). It seems probable that c o r r e l a t i o n of peaks with means i s an a r t i f a c t of open-top chamber experiments and the protocols used i n NCLAN because of the l i k e l i h o o d that chambers with the highest means w i l l a lso experience the highest peaks. Though the recent l i t e r a t u r e emphasizes the importance of peak concentrations (including the use of s t a t i s t i c s with thresholds) i n determining plant response, the exposure s t a t i s t i c s used i n the present study that take t h i s i n t o account (SUMXXX, IEXXX, HGTXXX P l [ l 2 ] , P l [ 7 ] , P12 and P7) d i d poorly as independent v a r i a b l e s i n the regressions employing the 1986 data (Tables 18-23). One might speculate that i t i s the sporadic nature with which the peaks occurred that leads to a poor r e l a t i o n s h i p . In t h i s experiment peak episodes above 100 ppb were rare (Appendix 3). Lefohn and Runeckles (1987) pointed out that under c e r t a i n circumstances the integrated exposure employed by Oshima (1976) may be appropriate where frequent high concentration exposures occur with l i t t l e time between them. This s i t u a t i o n de-emphasizes the need to consider recovery or repair between exposures. Three of the s t a t i s t i c s computed i n t h i s study took i n t o account the time at which peaks occurred i n the season. These were the number of days where the concentration exceeded 25 ppb, (D25x) f or the vegetative period, 174 and the periods of pod or tuber set and pod or tuber f i l l i n g ( t h i s could have been done for many of the exposure s t a t i s t i c s computed). These measures of 0 3 exposure d i d well for both peas and potatoes. Observations of the raw data (not presented) point to most occurrences of high peaks at the beginning of the season when plants were becoming established. At t h i s stage i t i s p o s s i b l e that peaks have l i t t l e l a s t i n g b i o l o g i c a l e f f e c t as young plants have generally been shown to be r e l a t i v e l y i n s e n s i t i v e (Menser et a l . , 1963; Reinert et a l . , 1972) and there i s ample time for recovery and r e p a i r . Exposures l a t e r i n the season during flowering or f r u i t set are l i k e l y to have more detrimental e f f e c t s on commercial y i e l d (Krupa and Teng, 1982 as c i t e d i n Krupa and K i c k e r t , 1987). The vegetative period appeared to produce the greatest response for peas, while the period of tuber f i l l i n g was most c r i t i c a l for potatoes. However, a s i n g l e season's data are i n s u f f i c i e n t to determine whether these r e s u l t s i n d i c a t e phenological stages of high s e n s i t i v i t y , or whether they merely r e f l e c t the pattern of exposures experienced during the 1986 season. Though NCLAN considers 25 ppb as the natural 0 3 background l e v e l i n much of the a g r i c u l t u r a l land i n the U.S. (with the exception of the South Coast A i r Basin of Southern C a l i f o r n a i ) , the ambient a i r p l o t s at U.B.C. ind i c a t e "normal" concentrations i n the range of 10-15 ppb during the summer months for 1986. Consequently, unlike the U.S. the numbers of days during which 1-h averages exceeded 25 ppb provides an i n d i c a t i o n of the days i n which 0 3 concentrations are elevated above the norm. As a group, the s t a t i s t i c s based upon t h i s c r i t e r i o n (D25, 2C25, 3C25 and 4C25) performed well for the 1986 data (Tables 18 to 23). The exposure s t a t i s t i c s were ranked in' terms of f i t (lowest RMS and 2 highest i ) for 1986 following the l i n e a r regressions for both crops (Table 175 24). This was done by ranking the performance of the exposure s t a t i s t i c s for each crop separately and then combining the ranks and taking the average for both crops. For 1986, the D25 1 performed the best o v e r a l l and HGT100 the worst. The arithmetic means (M7 and M12) ranked 11th and 12th out of 27 whereas the geometric mean (GM12) ranked 8th. Since the d i s t r i b u t i o n s of 0 3 concentrations achieved over the season were better described by lognormal than normal d i s t r i b u t i o n s , the geometric means more accurately characterize the data. Therefore i t i s not s u r p r i s i n g that they provided a better f i t to the data. The SUM100 ranked 24th out of 27 whereas t h i s was found to f i t best o v e r a l l by Lee et a l . (1987) using NCLAN and C o r v a l l i s Environmental Research Laboratory (CERL) data, obtained with open-top chambers. 7.5 Exposure-response functions The appropriate choice of a model whether simple l i n e a r , plateau-l i n e a r , Weibull or Box-Tidwell to name a few, i s s t i l l an unresolved issue i n the current l i t e r a t u r e . NCLAN rejected the use of l i n e a r models on the basis that they d i d not allow for curvature, and therefore no allowance for thresholds below which no y i e l d loss occurs was p o s s i b l e (USEPA, 1985). Though quadratic models allow for curvature and changes i n slope they cannot r e a s i l y accommodate thresholds i n response. Hence, the more f l e x i b l e , non-linear Weibull model has been used i n c r e a s i n g l y of l a t e , i n part because i t permits a point of i n f l e c t i o n i n the response curve (J.O. Rawlings, personal communication). Much of the a i r p o l l u t i o n l i t e r a t u r e s t i l l opts i n favor of simple l i n e a r regressions, where no threshold i n plant response i s apparent. In 176 Table 24. Overall combined ranking of the exposure s t a t i s t i c s used i n the linea r regression models for both crops i n 1986. Exposure s t a t i s t i c Rank 1 D25 1 2 D25 4 3 2C25 4 GH12 5 D25 2, H12 6 3C25, SUMALL 7 4C25 8 GM12 9 Ml[7] 10 D25 3 11 M12 12 M7, Ml[12]. 13 H10 14 GH10 15 D50 16 H14 17 Pl[7] 18 GH14 19 P7, P12 20 IE80 21 HRS 80 22 D80, SUM80 23 Pl[12], IE50 24 SUM100 25 IE100 26 SUM50 27 HRS100 177 the present study, l i n e a r regressions provided a good f i t to the 1986 y i e l d data. Though quadratic regressions were shown to improve the f i t , t h i s i s to be expected as they have greater f l e x i b l i t y (Tables 20, 21 and 23). Nevertheless the quadratic i s a symmetrical function and, as can be seen i n Figures 33 and 34, would eventually curve upwards, which makes l i t t l e b i o l o g i c a l sense at the higher O3 exposures. The Weibull function requires that the input data include the response to r e l a t i v e l y high p o l l u t a n t concentrations (causing 63% y i e l d l o s s ) , i n order to accurately e s t a b l i s h the shape of the response curve (Runeckles 1987; J.O. Rawlings personal communication). With O3 p o l l u t i o n t h i s concentration usually l i e s well outside the region of i n t e r e s t . Since such concentrations are much higher than most ambient exposure l e v e l s , they may t r i g g e r a d i f f e r e n t b i o l o g i c a l process from that causing y i e l d reductions at ambient l e v e l s . Consequently, curves generated from data where placement of the treatment l e v e l s i s inappropriate tend to have t h e i r greatest uncertainty at concentrations c l o s e s t to ambient (Runeckles 1987), which i s the range of i n t e r e s t i n present studies. Under these circumstances, the use of the Weibull function i n the present study was deemed inappropriate. Other approaches to developing comprehensive exposure response models have been more elaborate, such as Nosal 1s model outlined i n Section 2.4. Several of the terms used i n t h i s model were employed as exposure s t a t i s t i c s f o r the 1986 data, i . e . D25, M12I arid Pl[XX]. However, no attempt was made to use more elaborate functions i n the present study. Kohut et a l . (1987) u t i l i z e d l i n e a r , quadratic and Weibull models to characterize the r e l a t i o n s h i p between y i e l d of winter wheat cv. Vona and 178 ozone exposure. The best regression for the 1982 data was a quadratic function and a l i n e a r function i n 1983. Kress and M i l l e r (1983) a l s o found a s i g n i f i c a n t l i n e a r exposure-response r e l a t i o n s h i p f or soybean and had no evidence of a threshold i n response. Threshold responses are not evident i n the present study. Observations of Figures 37, 38 and 41 suggest that the data follow a concave rather than a convex shape. This has also been found by Kohut et a l . (1987) for winter wheat using a Weibull model. The shape of t h e i r exposure-response model was concave for 1982 data and convex for 1983. They suggested that these d i f f e r e n c e s i n shape were due to v a r i a t i o n s i n s o i l moisture status l i m i t i n g plant growth and decreasing uptake of O3, and a d i f f e r e n t length of exposure period for the two seasons. Foster et a l . (1983) i n a study of the e f f e c t s of O3 and S0 2 on tuber y i e l d and q u a l i t y of Centennial Russet potatoes, derived a l i n e a r regression using the SUMALL exposure s t a t i s t i c : Y i e l d = 1530 -15.8[SUMALL]. This i s comparable to r e s u l t s of the present study: Y i e l d = 1257.38 - 13.88[SUMALL] (Table 22, p. 154). At t h e i r highest exposure of 44.2 ppm.h the t o t a l tuber y i e l d was decreased by 45%. In the present study, a 42 + 14.09% reduction i n y i e l d was predicted at the highest exposure of 48.97 ppm.h using 19.10 ppm.h (Appendix 3) as the ambient a i r l e v e l . Stepwise multiple l i n e a r regression was used i n 1986 to tes t the r e l a t i v e contribution of various expressions of O3 exposure to changes i n y i e l d . Independent v a r i a t e s c h a r a c t e r i z i n g the number of episodes and the number of consecutive hours above a threshold contributed s i g n i f i c a n t l y to the f i n a l model for both pea and pod weights. Only the v a r i a t e f or episodes was s i g n i f i c a n t for potato tubers. Although a few of the 179 c o e f f i c i e n t s were p o s i t i v e the o v e r a l l r e l a t i o n s h i p between O3 and y i e l d was a negative one. The number of consecutive hours above a threshold encompassed the concept of the importance of previous exposures i n determining plant response. Though i s o l a t e d 1-h exposures occurring at any point i n the day may have a negative impact on y i e l d , i t was an t i c i p a t e d that a longer, low l e v e l exposure would allow recovery to take place, thus modifying the response seen, and perhaps r e s u l t i n g i n a number of the c o e f f i c i e n t s r e l a t e d to XC25 being p o s i t i v e . It should be pointed out that the l a s t three terms i n the stepwise regression, that i s , the XC25, are co-l i n e a r and hence, the estimated c o e f f i c i e n t s of these v a r i a t e s are unstable (USEPA, 1985). A negative l i n e a r and p o s i t i v e quadratic c o e f f i c i e n t found for many of the quadratic regressions i n 1986 (Tables 20, 21 and 23) suggests that y i e l d reduction i s not completely r e l a t e d to ambient concentration at the r e l a t i v e l y low l e v e l s experienced i n these studies. Even though the ambient concentration was increasing, various components of resistance to poll u t a n t uptake (Figure 1) may prevent s i m i l a r increases i n i n t e r n a l concentration. Hence y i e l d reduction i s not ne c e s s a r i l y l i n e a r l y r e l a t e d to increases i n ambient a i r p o l l u t a n t concentration. This once again r a i s e s the problem of attempting to r e l a t e y i e l d to ambient p o l l u t a n t concentration versus e f f e c t i v e dose. One of the uses of exposure-response information i s i n the set t i n g of a i r q u a l i t y standards. Chock (1982) has suggested that the c r i t e r i a for. s e t t i n g 0 3 standards should be based on a s t a t i s t i c not subject to large f l u c t u a t i o n s or large variance. He suggests that the 95th p e r c e n t i l e value of d a i l y maximum hourly concentrations would be appropriate. Use of high l e v e l unusual events to characterize exposure presents a problem since they 180 are u n r e l i a b l e estimates. Therefore, i t i s best to use le s s extreme values as they have les s tendancy to be autocorrelated, le s s uncertainty, and consequently are more r e l i a b l y obtainable from a i r q u a l i t y data. Using these c r i t e r i a , a number of the exposure s t a t i s t i c s computed from the present study such as the D25 x and GHXX would seem appropriate for use i n the setting of a i r q u a l t i y standards. F i n a l l y , the r e s u l t s obtained i n the present studies with peas and potatoes may be used to estimate the losses that may be occurring i n the Fraser V a l l e y as a r e s u l t of ozone p o l l u t i o n . For potatoes, losses of 39 + 14% are predicted by the regression of GH12 versus tuber fresh weight at 28.64 ppb 0^ (using 9.66 ppb as the ambient a i r l e v e l ) . For peas, losses of 61.34 + 4.59% are predicted by the regression of D25 2 versus pod fresh weight i n 1986, at 26 days of O3 above 25 ppb (using 4 episodes to t y p i f y the ambient a i r s i t u a t i o n ) . On the same basis, losses of 59.60 + 6.69% are predicted by the regression of D25 2 versus pea f r e s h weight i n 1986. It has not been possible to compute these exposure s t a t i s t i c s f o r d i f f e r e n t locations i n the Fraser V a l l e y during 1986. However, data from 1985 (supplied by the Greater Vancouver Regional D i s t r i c t ) showed the M7 to be 37 ppb at Chilliwack and 35 ppb at Abbotsford. These values (computed by Gordon Brown, U.B.C.) are s l i g h t l y lower than that of 48.7 ppb achieved i n the highest treatment i n 1985 and higher than the control value of 25.5 ppb. However, based on the 1986 experiment, losses of 43 + 15% are predicted by the regression of M7 versus tuber fresh weight at 37 ppb O3 (using 11.22 ppb as the ambient a i r l e v e l ) . Losses of 57 + 20% are predicted by the regression of M7 versus pod fresh weight and 50+ 25% for 181 ypea fresh weight at 37 ppb 0 3 (using 11.59 ppb as the ambient a i r s i t u a t i o n ) . The general agreement between the losses i n tuber fresh weight observed i n the present study and those observed by Foster et a l . , (1983) adds c r e d i b i l i t y to the former, and, since, the peas were grown under the same conditions as the potatoes i n the ZAPS p l o t s , suggests that the pea regressions are also probably r e a l i s t i c . If t h i s i s the case, the ambient ozone l e v e l s observed at various locations i n the Fraser V a l l e y are such as to suggest the occurrences of substantial losses of these two crops, peas and potatoes. The best case ( i e . the most conservative end of the confidence i n t e r v a l of the predicted losses) indicates that y i e l d reductions of 28% could be expected for peas and potatoes at present ambient ozone l e v e l s i n the Fraser V a l l e y . The worst case p r e d i c t s y i e l d reductions that are s u b s t a n t i a l l y greater. 182 8. SUMMARY 1. The e f f e c t of 0 3 on y i e l d of two c u l t i v a r s each of, carrots, peas, beans, b r o c c o l i and one c u l t i v a r of potato was investigated using a zonal a i r p o l l u t i o n system (ZAPS). 2. For both 1985 and 1986 the O3 concentration d i s t r i b u t i o n s achieved over the season were not normally d i s t r i b u t e d . They tended towards a lognormal d i s t r i b u t i o n . 3. A d d i t i o n a l analysis of the a i r q u a l i t y data from the ambient a i r p l o t (AA1) found other types of skewed d i s t r i b u t i o n s such as the 3-p Weibull, 3-p gamma and Johnsons S B (4-p log-normal) provided better d e s c r i p t i o n s of the data. 4. The concentration averaging time, the d a i l y time span over which the ozone concentrations were analyzed and the s e l e c t i o n c r i t e r i a used had a marked influence on the d i s t r i b u t i o n obtained. 5. The re l a t i o n s h i p s between 0 3 and y i e l d f o r both crops i n 1986 was an inverse one. 6. S i g n i f i c a n t l i n e a r reductions i n y i e l d were found for pea and pod fresh weight using the D25 2 exposure s t a t i s t i c i n 1986. 7. A s i g n i f i c a n t l i n e a r reduction i n y i e l d was found for potato tuber fresh weight using the GH12 exposure s t a t i s t i c i n 1986. 8. A s i g n i f i c a n t multiple l i n e a r regression was found for pea fre s h weight using D25, 2C25, 3C25 and 4C25 as independent v a r i a t e s and for pod fresh weight using D25, 2C25, and 3C25 as independent v a r i a t e s . 9. The D25 1 exposure s t a t i s t i c performed best i n 1986, across both crops for the l i n e a r regressions. 10. The ZAPS has been shown to be a useful t o o l for exposing crops to randomly f l u c t u a t i n g ozone concentrations i n a f i e l d s i t u a t i o n . 183 11. Based on the 1986 experiment the best case estimate indicates that y i e l d reductions of 28% could be expected for peas and potatoes at 37 ppb ozone (expressed as the season-long 7-h mean, M7). 12. A number of the new exposure s t a t i s t i c s were shown to provide good f i t s to the y i e l d data (D25 x and GHXX). These s t a t i s t i c s are e a s i l y c a l c u lated from ambient a i r data and may prove to be v i a b l e a l t e r n a t i v e s to those exposure s t a t i s t i c s c u r r e n t l y i n use, for assessing the p o t e n t i a l impact of ozone on crops i n the Fraser V a l l e y and the development of a i r q u a l i t y standards. 184 Adomait, E.J., J . Ensing and G. Hofstra. 1987. A dose-response function for the impact of ozone on Ontario grown white bean and an estimate of economic l o s s . Can. J . Plant S c i . 67: 131-136. Amiro, B.D., T.J. G i l l e s p i e and G.W. T h u r t e l l . 1984. Injury response of Phaseolus v u l g a r i s to ozone f l u x density. Atmos. Environ. 18: 1207-1215. Amundson, R.G., R.M. Raba, A.W. Schoettle and P.B. Reich. 1986. Response of soybean to low concentrations of ozone: I I . E f f e c t s on growth, biomass a l l o c a t i o n , and flowering. J . Environ. Qual. 15_: 161-167. Ashenden, T.W., and T.A. Mansfield. 1977. Influence of wind speed on the s e n s i t i v i t y of Ryegrass to S0 2- J . Expt. Bot. 28: 729-735. Baldocchi, D.D., B.B. Hicks and P. Camara. 1987. A canopy stomatal resistance model for gaseous deposition to vegetated surfaces. Atmos. Environ. 21: 91-101. Barclay, G.M., H.J. Murphy, F.E. Manzer and F.E. Hutchinson. 1973. E f f e c t s of d i f f e r e n t i a l rates of nitrogen and phosphorus on early b l i g h t i n potatoes. Am. Pot. J . 50: 42-48. Bencala, K.E. and J.H. S e i n f e l d . 1976. On frequency d i s t r i b u t i o n s of a i r p o l l u t a n t concentrations. Atmos. Environ. 10: 941-950. Bennett, J.P., A. Clyde H i l l and D.M. Gates. 1973. A model for gaseous pol l u t a n t sorption by leaves. JAPCA. 11: 957-962. Bennett, J.P., H.M. Resh and V.C. Runeckles. 1974. Apparent stimulations of plant growth by a i r p o l l u t a n t s . Can. J . Bot. 52: 35-42 Bennett, J.P. and R.J. Oshima. 1976. Carrot i n j u r y and y i e l d response to ozone. J . Amer. Soc. Hort. S c i . 101: 638-639. Bennett, J.P. and V.C. Runeckles. 1977. E f f e c t s of low l e v e l s of ozone on growth of crimson clover and annual ryegrass. Crop Science. 17: 443-445. Bisessar, S. 1982. E f f e c t of ozone, antioxidant protection, and early b l i g h t on potato i n the f i e l d . J . Amer. Soc. Hort. S c i . 107: 597-599. Blum, U.D.O., E. Mrozek J r . and E. Johnson. 1983. Investigation of ozone (0 3) e f f e c t s on 1 4 C d i s t r i b u t i o n i n Ladino clover. Environ, and Expt. Bot. 23: 369-378. Brennan, E. and A. Rhoads. 1976. Response of field-grown bean c u l t i v a r s to atmospheric oxidant i n New Jersey. Plant Disease Reporter. 60: 941-945. 185 Buttazzoni, C , I. Lavagnini, A. Marani, F. Z i l l o Grandi and A. Del Turco. 1986. P r o b a b i l i t y model for atmospheric sulphur dioxide concentrations i n the area of Venice. JAPCA. 36: 1028-1030. Carnahan, J.E., E.L. Jenner and E.K.W. Wat. 1978. Prevention of ozone i n j u r y to plants by a new protectant chemical. Phytopathology. 68: 1225-1229. Chock, D.P. and P.S. Sluchak. 1986. Estimating extreme values of a i r q u a l i t y data using d i f f e r e n t f i t t e d d i s t r i b u t i o n s . Atmos. Environ. 20: 989-993. Chung, Y.S. and T. Dann. 1985. Observations of stratospheric ozone at the ground l e v e l i n Regina, Canada. Atmos. Environ. 19: 157-162. Clarke, B.B., M. R. Henninger and E. Brennan. 1983. An Assessment of potato losses caused by oxidant a i r p o l l u t i o n i n New Jersey. Phytopathology. 73: 104-108. Colbeck, I. and R.M. Harrison. 1985. The frequency and causes of elevated concentrations of ozone at ground l e v e l at r u r a l s i t e s i n N.W. England. Atmos. Environ. 19: 1577-1587. Cure, W.W., J.S. Sanders and A.S. Heagle. 1986. Crop y i e l d response predicted with d i f f e r e n t characterizations of the same ozone treatments. J . Environ. Qual. 15: 251-254. Curran, T.C. and J.C. Suggs. 1986. E f f e c t s of measurement uncertainty on a i r q u a l i t y summary s t a t i s t i c s . Atmos Environ. 20: 571-576. Davis, D.D. and L. Kress. 1974. The r e l a t i v e s u s c e p t i b i l i t y of ten bean v a r i e t i e s to ozone. Plant Disease Reporter. 58: 14-16. DeVos, N.E., E.J. P e l l , R.R. H i l l , J r . , and R.H. Cole. 1983. Laboratory versus f i e l d response of potato gerotypes to oxidant s t r e s s . Plant Disease. 67: 173-176. Dugger, W.M., J r . , O.C. Taylor, Eugene C a r d i f f and C.R. Thompson. 1962. Stomatal action i n plants as r e l a t e d to damage from photochemical oxidants. Plant P h y s i o l . 37: 487-491. Dugger, W.M., J r . , O.C. Taylor, Eugene C a r d i f f and C. Ray Thompson. 1962. Relationship between carbohydrate content and s u s c e p t i b i l i t y of pinto bean plants to ozone damage. J . Amer. Soc. for Hort. S c i . 81: 304-315. Dunning, J.A. and W.W. Heck. 1973. Response of pinto bean to ozone as conditioned by l i g h t i n t e n s i t y and/or humidity. Environ. S c i . and Technol. 17: 824-826. Dunning, J.A. and W.W. Heck. 1977. Response of bean and tobacco to ozone: e f f e c t of l i g h t i n t e n s i t y , temperature and r e l a t i v e humidity. JAPCA. 27: 882-886. \ 186 Engle, R.L. and W.H. Gabelman. 1967. The e f f e c t s of low l e v e l s of ozone on pinto beans, Phaseolus v u l g a r i s L. J . Amer. Soc. Hort. S c i . 91; 304-309. Evans, G., P. F i n k e l s t e i n , B. Martin, N. P o s s i e l and M. Graves. 1983. Ozone measurements from a network of remote s i t e s . JAPACA. 33: 291-296. Evans, L.S. and I.P. Ting. 1973. Ozone-induced membrane permeability changes. Amer. J . Bot. 60: 155-162. Foster, K.W., J.P. Guerard, R.J. Oshima, J.C. Bishop and H. Timm. 1983a. D i f f e r e n t i a l ozone s u s c e p t i b i l i t y of centennial russet and white rose potato as demonstrated by fumigation and antioxidant treatments. Am. Pot. Journal. 60: 127-139. Foster, K.W., H. Timm, C K . Labanauskas and R.J. Oshima. 1983b. E f f e c t s of ozone and sulphur dioxide on tuber y i e l d and q u a l i t y of potatoes. J . Environ. Qual. 12: 75-80. Fowler D., and J.N. Cape. 1982. A i r pollutants i n a g r i c u l t u r e and h o r t i c u l t u r e . In E f f e c t s of Gaseous A i r P o l l u t i o n i n A g r i c u l t u r e and  H o r t i c u l t u r e (edited by Unsworth M.H. and Ormrod D.P.) pp3-26. Butterworth S c i e n t i f i c , London. Georgopoulos, P.G. and J.H. S e i n f e l d . 1982. S t a t i s t i c a l d i s t r i b u t i o n s of a i r p o l l u t a n t concentrations. Environ. S c i . Technol. 16: 401A-416A. Haas, J.H. 1970. Relation of crop maturity and physiology to a i r p o l l u t i o n i n c i t e d bronzing of Phaseolus v u l g a r i s . Phytopathology. 60: 407-410. Heagle, A.S., W.W. Heck and D. Body. 1971. Ozone i n j u r y to plants as influenced by a i r v e l o c i t y during exposure. Phytopathology. 61: 1209-1212. Heagle, A.S., D.E. Body and E.K. Pounds. 1972. E f f e c t of ozone on y i e l d of sweet corn. Phytopathology. 62: 683-687. Heagle, A.S., D.E. Body and W.W. Heck. 1973. An open-top f i e l d chamber to assess the impact of a i r p o l l u t i o n on plants. J . environ. Qual. 2: 365-368. Heagle, A.S., R.B. Philbeck, H.H. Rogers and M.B. Letchworth. 1979a. Dispensing and monitoring ozone i n open-top f i e l d chambers for p l a n t -e f f e c t s studies. Phytopathology. 69: 15-20. Heagle, A.S., R.B. Philbeck and W.M. Knott. 1979b. Thresholds for i n j u r y , growth, and y i e l d loss caused by ozone on f i e l d corn hybrids. Phytopathology. 69: 21-26. Heagle, A.S., S. Spencer and M.B. Letchworth. 1979c. Y i e l d response of winter wheat to chronic doses of ozone. Can. J . Bot. 57: 1999-2005. 187 Heagle, A.S., R.B. Philbeck and M.B. Letchworth. 1979d. Injury and y i e l d responses of spinach c u l t i v a r s to chronic doses of ozone i n open-top f i e l d chambers. J . Environ. Qual. 8: 368-373. Heagle, A.S., M.B. Letchworth and C.A. M i t c h e l l . 1983a. E f f e c t s of growth medium and f e r t i l i z e r rate on the y i e l d response of soybeans exposed to chronic doses of ozone. Phytopathology. 73: 134-139. Heagle, A.S., W.W. Heck, J.O. Rawlings and R.B. Philbeck. 1983b. E f f e c t s of chronic doses of ozone and sulphur dioxide on i n j u r y and y i e l d of soybeans i n open-top f i e l d chambers. Crop Science. 23: 1184-1191. Heagle, A.S., V.M. Lesser, J.O. Rawlings, W.W. Heck and R.B. Philbeck. 1986. Response of soybeans to chronic doses of ozone applied as constant or proportional additions to ambient a i r . Phytopathology. 76: 51-56. Heagle, A.S., W.W. Heck, V.M. Lesser and J.O. Rawlings. 1987. E f f e c t s of d a i l y ozone exposure duration and concentration f l u c t u a t i o n on y i e l d of tobacco. Phytopathology. 77: 856-862. Heck, W.W., J.A. Dunning and I . J . Hindawi. 1965. Interactions of environmental factors on the s e n s i t i v i t y of plants to a i r p o l l u t i o n . JAPCA. 15: 511-515. Heck, W.W., J.A. Dunning and I . J . Hindawi. 1966. Ozone: non l i n e a r r e l a t i o n of dose and i n j u r y i n p l a n t s . Science. 151: 577-578. Heck, W.W., O.C. Taylor, R. Adams, G. Bingham, J . M i l l e r , E. Preston and L. Weinstein. 1982. Assessment of crop loss from ozone. JAPCA. 32: 353-361. Heck, W.W., R.M. Adams, W.W. Cure, A.S. Heagle, H.E. Heggestad, R.J. Kohut, L.W. Kress, J.O. Rawlings and O.C. Taylor. 1983. A reassessment of crop los s from ozone. Environ. S c i . Technol. 17: 573A-581A. Heck, W.W., W.W. Cure, J.O. Rawlings, L.J. Zaragoza, A.S. Heagle, H.E. Heggestad, R.J. Kohut, L.W. Kress and P.J. Temple. 1984a. Assessing impacts of ozone on a g r i c u l t u r a l crops. I. Overview. JAPCA. 34: 729-735. Heck, W.W., W.W. Cure, J.O. Rawlings, L.J. Zaragoza, A.S. Heagle, H.E. Heggestad, R.J. Kohut, L.W. Kress and P.J. Temple. 1984b. Assessing impacts of ozone on a g r i c u l t u r a l crops: I I . Crop y i e l d functions and a l t e r n a t i v e exposure s t a t i s t i c s . JAPCA. 34: 810-817. Heggestad, H.E. 1973. Photochemical a i r p o l l u t i o n i n j u r y to potatoes i n the A t l a n t i c coastal states. Am. Pot. Jour. 50: 315-329. Heggestad, H.E., A.S. Heagle, J.H. Bennett and E.J. Koch. 1980. The e f f e c t s of photochemical oxidants on the y i e l d of snap beans. Atmos. Environ. 14: 317-326. 188 Heggestad, H.E., T.J. Gish, E.H. Lee, J.H. Bennett and L.W. Douglas. 1985. Interaction of s o i l moisture stress and ambient ozone on growth and y i e l d s of soybeans. Phytopathology. 75: 472-477. Henderson, W.R. and R.A. Reinert. 1979. Y i e l d response fo four fresh market tomato c u l t i v a r s a f t e r acute ozone exposure i n the seedling stage. J . Amer. Soc. Hort. S c i . 104: 754-759. H i l l , A.C, M.R. Pack, M. Treshow, R.J. Downs and L.G. Transtrum. 1961. Plant i n j u r y induced by ozone. Phytopathology. 51: 356-363. H i l l , A.C. and N. L i t t l e f i e l d . 1969. Ozone. E f f e c t on apparent photosynthesis, rate of t r a n s p i r a t i o n , and stomatal closure i n plants. Environ. S c i . and Technol. 3: 52-56. Hogsett, W.E., D.T. Tingey and S.R. Holman. 1985. A programmable exposure control system for determination of the e f f e c t s of pollutant exposure regimes on plant growth. Atmos. Environ. 19: 1135-1145. Holland, D.M. and T. Fitz-Simons. 1982. F i t t i n g s t a t i s t i c a l d i s t r i b u t i o n s to a i r q u a l i t y data by the maximum l i k e l i h o o d method. Atmos. Environ. 16: 1071-1076. Holley, J.D., G. Hofstra and R. H a l l . 1985. E f f e c t of reducing oxidant i n j u r y and ea r l y b l i g h t on fresh weight and tuber density of potato. Phytopathology. 75: 529-532. Hooker, W.J., T.C. Yang and H.S. Potter. 1973. A i r p o l l u t i o n i n j u r y of potato i n Michigan. Am. Pot. Jour. 50: 151-161. Hucl, P. and W.D. Beversdorf. 1982. The response of selected Phaseolus  v u l g a r i s L. c u l t i v a r s to ozone under c o n t r o l l e d fumigation and ambient f i e l d l e v e l s . Can. J . Plant S c i . 62: 561-569. Jacobson, J.S. 1982. Ozone and and the growth and p r o d u c t i v i t y of a g r i c u l t u r a l crops. In E f f e c t s of Gaseous A i r P o l l u t i o n i n  Agr i c u l t u r e and Hor t i c u l t u r e (edited by Unsworth M.H. and Ormrod D.P.) pp293-304. Butterworth S c i e n t i f i c , London. James, R.L., F.W. Cobb J r . , P.R. M i l l e r and J.R. Parameter J r . 1980a. E f f e c t s of oxidant a i r p o l l u t i o n on s u s c e p t i b i l i t y i f pine roots to Fomes annosus. Phytopath. 70: 560-563. Johnston, J.W., J r . and A.S. Heagle. 1982. Response of c h r o n i c a l l y ozonated soybean plants to an acute ozone exposure. Phytopathology. 72: 387-389. K e l l y , N.A., G.T. Wolff and M.A. Ferman. 1984. Sources and sinks of ozone i n r u r a l areas. Atmos. Environ. 18: 1251-1266. Kobriger, J.M., T.W. T i b b i t t s and M.L. Brenner. 1984. Injury, stomatal conductance and a b s c i s i c a c i d l e v e l s of pea plants following ozone plus sulfur dioxide exposures at d i f f e r e n t times of the day. Plant Physiol. 76: 823-826. 189 Kobriger, J.M. and T.W. T i b b i t t s . 1985. E f f e c t of r e l a t i v e humidity p r i o r to and during exposure of peas to ozone and sulphur dioxide. J . Amer. Soc. Hort. S c i . 110: 21-24. Kohut, R.J., R.G. Amundson, J.A. Laurence, L. Colavito, P. van Leuken and P. King. 1987. E f f e c t s of ozone and sulphur dioxide on y i e l d of winter wheat. Phytopathology. 77: 71-74. K o z i o l , M.J. 1980. Monitoring gas concentrations i n p o l l u t a n t exposure systems: def i n i n g exposure concentrations. J . Expt. Bot. 31: 1413-1423. Kress, L.W. and J.E. M i l l e r . 1983. Environ. Qual. 12: 276-281. Kress, L.W. and J.E. M i l l e r . 1984. Can. J . Bot. 63: 2408-2415. Impact of ozone on soybean y i e l d . J . Impact of ozone on f i e l d - c o r n y i e l d . Kress, L.W., J.E. M i l l e r and H.J. Smith. 1985. Impact of ozone on winter wheat y i e l d . Environ, and Expt. Bot. 25: 211-288. Krupa, S. and R.N. Kickert. 1987. An analysis of numerical models of a i r po l l u t a n t exposure and vegetation response. Environ. P o l l . 44: 127-158. Krupa, S., R.N. Kickert and M. Nosal. 1986. An analysis of numerical models of a i r pollutant exposure and vegetation response. In Proceedings of Second Alberta Symposium: a c i d forming emissions i n Alberta and t h e i r e c o l o g i c a l e f f e c t s . May 12-15. Calgary, Alberta. Krupa, S. 1984. F i e l d exposure methodology for assessing the e f f e c t s of photochemical oxidants on crops. In: Proc. 77th Ann. Meeting of the  A i r P o l l u t i o n Control Association, San Francisco, CA. Larsen, R.I. 1973. An a i r q u a l i t y data analysis system for i n t e r r e l a t i n g e f f e c t s standards and needed source reductions. JAPCA. 23: 933-940. Larsen, R.I. and W.W. Heck. 1984. An a i r q u a l i t y data a n a l y s i s system for i n t e r r e l a t i n g e f f e c t s , standards, and needed source reductions: Part 8. An e f f e c t i v e mean ozone crop reduction mathematical model. JAPCA. 34: 1023-1034. Laurence, J.A. 1981. E f f e c t s of a i r p o l l u t a n t s on plant-pathogen i n t e r a c t i o n s . Z. Pflanzenkr. Pflahzenschutz 87: 156-172. Lee, J . , R. Lewis and D. Body. 1975. A f i e l d experimental system for the evaluation of the bioenvironmental e f f e c t s of sulphur dioxide. Montana: Fort Union Coal F i e l d Symposium. Academy of Science. 190 Lee, J . J . , E.M. Preston and R.A. Lewis. 1977. A system for the experimental evaluation of the e c o l o g i c a l e f f e c t s of sulphur dioxide. Fourth J o i n t Conf. on Sensing of Environmental P o l l u t a n t s . Am. Chem. Soc. Lee, J . J . and R.A. Lewis. 1977. Zonal a i r p o l l u t i o n system design and performance. In Bioenvironmental Impact of a C o a l - f i r e d Power Plant. Section 9. C o l s t r i p , Montana: EPA t h i r d i n terim report. Lefohn, A.S. and H.M. Benedict. 1982. Development of a mathematical index that describes ozone concentration, frequency and duration. Atmos. Environ. 16: 2529-2532. Lefohn, A.S. and D.T. Tingey. 1985. Discussion of paper: Injury response of Phaseolus v u l g a r i s to ozone f l u x density (Atmos. Environ. 18: 1207-1215). Atmos. Environ. 19: 206-208. Lefohn A.S. and V.C. Runeckles. 1987. Es t a b l i s h i n g standards to protect vegetation-ozone expoosure/dose considerations. Atmos. Environ. 21: 561-568. Lefohn,A.S., A.A. Laurence and R.J. Kohut. 1988. A comparison of i n d i c i e s that describe the r e l a t i o n s h i p between exposure to ozone and reduction i n the y i e l d of a g r i c u l t u r a l crops. Atmos. Environ. Leone, I.A. and E. Brennan. 1970. Ozone t o x i c i t y i n tomato as modified by phosphorus n u t r i t i o n . Phytopathology. 60: 1521-1524. Leone, I.A. 1976. Response of potassium-deficient tomato plants to atmospheric ozone. 1976. Phytopathology. 66: 734-736. Lewis, E. and E. Brennan. 1977. A d i s p a r i t y i n the ozone response of bean plants grown i n a greenhouse, growth chamber or open-top chamber. JAPCA. 27: 889-891. Luria, M., H. Almog and M. Peleg. 1984. Transport and transformation of a i r p o l l u t a n t s from I s r a e l s coastal area. Atmos. Environ. 18: 2215-2221. Macdowall, F.D.H., E.I. Mukammal and A.F.W. Cole. 1964. Direct c o r r e l a t i o n of a i r - p o l l u t i n g ozone and tobacco weather f l e c k . Can. J . Plant S c i . 44: 410-417. Male, L.M. 1982. An experimantal method for p r e d i c t i n g plant y i e l d response to p o l l u t i o n time s e r i e s . Atmos. Environ. 16: 2247-2252. Male, L., E. Preston and G. Neely. 1983. Y i e l d response curves of crops exposed to sulphur dioxide time s e r i e s . Atmos. Environ. 17: 1589-1593. Mancini J.L. 1983. A method for c a l c u l a t i n g e f f e c t s , on aquatic organisms, of time varying concentrations. Water Res. 17: 1355-1362. 191 Mandl, R.H., L.H. Weinstein, D.C. McCune and M. Kevenv. 1973. A c y l i n d r i c a l , open-top chamber for the exposure of plants to a i r po l l u t a n t s i n the f i e l d . J . Environ. Qual. 2: 371-376. Manning, W.J., W.A. Feder, I. Perkins and M. Glickman. 1969. Ozone i n j u r y and i n f e c t i o n of potato leaves by B o t r y t i s cinerea. Plant Dis. Rep. 53: 691-693. Marie, B.A. and D.P. Ormrod. 1986. Dose response relationshps of the growth and i n j u r y e f f e c t s of ozone and sulphur dioxide on Brassicaceae seedlings. Can. J . Plant S c i . 6_6: 659-667. Medeiros, W.H. and P.D. Moskowitz. 1983. Quantifying e f f e c t s of oxidant a i r pollutants on a g r i c u l t u r a l crops. Environmental International. 9: 505-513. Meiners, J.P. and H.E. Heggestad. 1979. -Evaluation of snap bean c u l t i v a r s for resistance to ambient oxidants i n f i e l d p l o t s and to ozone i n chambers. Plant Disease Reporter. 63: 273-277. Menser, H.A., H.E. Heggestad and O.E. Street. 1963. Response of plants to a i r p o l l u t a n t s . I I . E f f e c t s of ozone concentration and leaf maturity on i n j u r y to Nicotiana tabacum. Phytopath. 53: 1304-1308. Mosley, A.R., R.C. Rowe and T.C. Weidensaul. 1978. Relationship of f o l i a r ozone i n j u r y to maturity c l a s s i f i c a t i o n and y i e l d of potatoes. Am. Pot. Jour. 55: 147-153. Mukammal, E.I. 1965. Ozone as a cause of tobacco i n j u r y . Agr. Meteorol. 2: 145-165. Musselman, R.C, R.J. Oshima and R.E. Gallavan. 1983. S i g n i f i c a n c e of pol l u t a n t concentration d i s t r i b u t i o n i n the response of red kidney' beans to ozone. J . Amer. Soc. Hort. S c i . 108: 347-351. Musselman, R.C, A.J. Huerta, P.M. McCool and RJ. Oshima. 1986. Response of beans to simulated ambient and uniform ozone d i s t r i b u t i o n with equal peak concentrations. J . Amer. Hort. S c i . I l l : 470-473. McLaughlin, S.B. and G.E. Taylor. 1981. Relative humidity: important modifier of po l l u t a n t uptake by p l a n t s . Science. 211: 167-168. McLeod, A.R., J.E. F a c k r e l l and K. Alexander. 1985. Open-air fumigation of f i e l d crops: c r i t e r i a and design f o r a new experimental system. Atmos. Environ. 19: 1639-1649. Nouchi, I. and K. Aoki. 1979. Morning glory as a photochemical oxidant i n d i c a t o r . Environ. P o l l u t . 18: 289-303. Nosal, M. 1984. S t a t i s t i c a l models f o r a i r pollutant dose and plant response. San Francisco, C a l i f o r n i a : 77th Annual Meeting of A i r P o l l u t i o n Control Association. 192 O'Dell, R.A., M. Taheri and R.L. Kabel. 1977. A model for uptake of p o l l u t a n t s by vegetation. JAPCA. 27: 1104-1109. Okano, K., 0. Ito, G. Takeba, A. Shimizu and T. Totsuka. 1984. A l t e r a t i o n of " C - a s s i m i l a t e p a r t i t i o n i n g i n plants of Phaseolus v u l g a r i s exposed to ozone. New Phytol. 97: 155-163. Olszyk, D.M., T.W. T i b b i t t s and W.M. Hertzberg. 1980. Environment i n open-top chambers u t i l i z e d for a i r p o l l u t i o n studies. J . Environ. Qual. 9: 610-615 Olszyk, D.M. and D.T. Tingey. 1984. Phytotoxicity of a i r pollutants -evidence for the photodetoxification of sulphur dioxide but not ozone. Plant Physiol. 74: 999-1005. Olszyk, D.M. and D.T. Tingey. 1985. Metabolic basis for i n j u r y to plants from combinations of ozone and sulphur dioxide. Plant Physiol. 77: 935-939. Ormrod, D.P. 1976. S e n s i t i v i t y of pea c u l t i v a r s to ozone. Plant Disease Reporter. 60: 423-426. Oshima, R.J. 1973. E f f e c t of ozone on a commercial sweet corn v a r i e t y . Plant Disease Reporter. 57: 719-723. Oshima, R.J., O.C. Taylor, P,K. Braegelmann and D.W. Baldwin. 1975. E f f e c t of ozone on the y i e l d and plant biomass of a commercial v a r i e t y of Tomato. J . Environ. Qual. 4: 463-464. Oshima, R.J., M.P. Poe, P.K. Braegelmann, D.W. Baldwin and V. VanWay. 1976. Ozone dosage-crop loss function for a l f a l f a : A standardized method for assessing crop losses from a i r p o l l u t a n t s . JAPCA. 26: 861-865. Oshima, R . J i f P.K. Braegelmann, D.W. Baldwin, V.V. Way and O.C. Taylor. 1977a. Responses of f i v e c u l t i v a r s of fresh market tomato to ozone: A contrast of c u l t i v a r screening with f o l i a r i n j u r y and y i e l d . J . Amer. Soc. Hort. S c i . 102: 286-289. Oshima, R.J., P.K. Braegelmann, D.W. Baldwin, V.V. Way and O.C. Taylor. 1977b. Reduction of tomato f r u i t s i z e and y i e l d by ozone. J . Amer. Soc. Hort. S c i . 102: 289-293. Oshima, R.J., J.P. Bennett and P.K. Braegelmann. 1978. E f f e c t of ozone on growth and a s s i m i l a t e p a r t i t i o n i n g i n parsley. J . Amer. Soc. Hort. S c i . 103: 348-350. Oshima, R.J., P.K. Braegelmann, R.B. F l a g l e r and R.R. Teso. 1979. The e f f e c t s of ozone on the growth, y i e l d and p a r t i t i o n i n g of dry matter i n cotton. J . Environ. Qual. 8: 474-479. Otto, H. and R.H. Daines. 1969. Plant i n j u r y by a i r p o l l u t a n t s : influence of humidity on stomatal apertures and plant response to ozone. Science. 163: 1209-1210. 193 P e l l , E.J. and N.S. Pearson. 1984. Ozone-induced reduction i n quantity and q u a l i t y of two potato c u l t i v a r s . Environ. P o l l . 35: 345-352. Pratt, G.C., R.C. Hendrickson, B.I. Chevone, D.A. Christopherson, M.V. O'Brien and S.V. Krupa. 1983. Atmos. Environ. 17: 2013-2023. Rajput, C S . and D.P. Ormrod. 1976. Response of eggplant c u l t i v a r s to ozone. Hort Science. 11: 462-463. Rawlings, J.O. and W.W. Cure. 1985. The Weibull function as a dose-response model to describe ozone e f f e c t s on crop y i e l d s . Crop Science. 25: 807-814 Reich, P.B. and R.G. Amundson. 1984. Low l e v e l ozone and/or sulphur dioxide exposure causes a l i n e a r d e c l i n e i n soybean y i e l d . Environ. P o l l . 34: 345-355. Reich, P.B., A..W. Schoettle and R.G. Amundson. 1985. E f f e c t s of low concentrations of ozone, leaf age and water stress of leaf d i f f u s i v e conductance and water use e f f i c i e n c y i n soybean. P h y s i o l . Plant. 63: 58-64. Reinert, R.A., D.T. Tingey and H.B. Carter. 1972a. S e n s i t i v i t y of tomato c u l t i v a r s to ozone. J . Amer. Soc. Hort. S c i . 97: 149-151. Reinert, R.A., D.T. Tingey and H.B. Carter. 1972b. Ozone induced f o l i a r i n j u r y i n l e t t u c e and radish c u l t i v a r s . J . Amer. Soc. Hort. S c i . 97: 711-714. Reinert, R.A. and P.V. Nelson. 1979. S e n s i t i v i t y and growth of tweleve e l a t i o r begonia c u l t i v a r s to ozone. Hort. S c i . 14: 747-748. Reinert, R.A. and W.R. Henderson. 1980. F o l i a r i n j u r y and growth of tomato c u l t i v a r s as influenced by ozone dose and plant age. J . Amer. Soc. Hort. S c i . 105: 322-324. Reinert, R.A., J.A. Dunning, W.W. Heck, P.S. Benepal and M. Rangappa. 1984. Screening of bean (P. v u l g a r i s ) f or s e n s i t i v i t y to ozone. Hort Science. 19: 86-88. Rich, S. and N.C Turner. 1972. Importance of moisture on stomatal behavior of plants subjected to ozone. JAPCA. 22: 718-721. Roberts, T.M. 1984. E f f e c t s of a i r p o l l u t a n t s on a g r i c u l t u r e and f o r e s t r y . Atmos. Environ. 18: 629-652. Runeckles, V.C. 1974. Dosage of a i r p o l l u t a n t s and damage to vegetation. Environ. Conservation. 1: 305-308. Runeckles, V.C and P.M. Rosen. 1974. E f f e c t s of pretreatment with low ozone concentrations on ozone i n j u r y to bean and mint. Can. J . Bot. 52: 2607-2610. 194 Runeckles, V.C. and P.M. Rosen. 1977. E f f e c t s of ambient ozone pretreatment on t r a n s p i r a t i o n and s u s c e p t i b i l i t y to ozone i n j u r y . Can. J . Bot. 55: 193-197. Runeckles, V.C. and K.T. Palmer. 1981a. Design and p h y s i c a l c h a r a c t e r i s t i c s of amodified Zonal A i r P o l l u t i o n System for studying sulphur dioxide e f f e c t s on vegetation. Paper presented at P a c i f i c Northwest International Secrion Meeting, A i r P o l l u t i o n Control Assoc., Edmonton, A l b e r t a . Runeckles, V.C, K.T. Palmer and H. T r a b e l s i . 1981b. E f f e c t s of f i e l d exposures to sulphur dioxide on douglas f i r Agropyron spicatum and Lolium perenne. S i l v a Fennica. 15: 405-415. Runeckles, V.C. 1987. Exposure, dose, vegetation response and standards: w i l l they ever be related? Proc. annual meeting of the A i r  P o l l u t i o n Control Assoc., Paper 87-33.2. Pottsburgh, PA, A i r P o l l u t i o n Control Assoc. (1987). Runeckles, V.C. 1988. Comment on a comparison of indic e s that describe the r e l a t i o n s h i p between exposure to ozone and reduction i n the y i e l d of a g r i c u l t u r a l crops. Atmos. Environ. Smith, G., B. Greenhalgh, E. Brennan and J . J u s t i n . 1987. Soybean y i e l d i n New Jersey r e l a t i v e to ozone p o l l u t i o n and antioxidant a p p l i c a t i o n . Plant Disease. 71: 121-125. Steele, R.G.D. and J.H. T o r r i e . 1980. P r i n c i p l e s and procedures of  s t a t i s t i c s . Second e d i t i o n . McGraw-Hill, Inc., U.S. Taylor, G.E., D.T. Tingey and H.C Ratsch. 1982. Ozone f l u x i n Glycine max (L.) Merr.: Sit e s of regulation and r e l a t i o n s h i p of leaf i n j u r y . Oecologia ( B e r l ) . 53: 179-186. Taylor, J.A., A.J. Jakeman and R.W. Simpson. 1986. Modelling d i s t r i b u t i o n s of a i r p o l l u t a n t concentrations. I. I d e n t i f i c a t i o n of s t a t i s t i c a l models. Atmos. Environ. 20: 1781-1789. Temple, P.J., 0. C l i f t o n Taylor and L.F. Benoit. 1985. Cotton y i e l d responses to ozone as mediated by s o i l moisture and evapotranspiration. J . Environ. Qual. 14: 55-59. Temple, P.J., 0. C. Taylor and L.F. Benoit. 1986. Y i e l d resposne of head l e t t u c e (Latuca s a t i v a L.) to ozone. Environ, and Expt. Bot. 26: 53-58. Ting, I.P. and W.M. Dugger, J r . 1968. Factors a f f e c t i n g ozone s e n s i t i v i t y and s u s c e p t i b i l i t y of cotton plants. JAPCA. 18: 810-813. Ting, I.P. and S.K. Mukerji. 1971. leaf ontogeny as a f a c t o r i n s u s c e p t i b i l i t y to ozone: amino a c i d and carbohydrate changes during expansion. Amer. J . Bot. 58: 497-504. 195 Tingey, D.T., W.W. Heck and R.A. Reinert. 1971a. E f f e c t of low concentrations of ozone and sulphur dioxide on f o l i a g e , growth and y i e l d of rad i s h . J . Amer. Soc. Hort. S c i . 96: 369-371. Tingey, D.T. and W.E. Hogsett. 1985. Water stress reduces ozone i n j u r y v i a a stomatal mechanism. Plant Physiol. 77: 944-947. Turner, N.C., S. Rich and P.E. Waggoner. 1973. Removal of ozone by s o i l . J . Environ. Qual. 2: 259-263. Turner, N.C., P.E. Waggoner and S. Rich. 1974. Removal of ozone from the atmosphere by s o i l and vegetation. Nature. 250: 486-489. USEPA. 1985. A i r Quality c r i t e r i a f o r ozone and other photochemical oxidants. V o l. 3. pp 7-1-7-297. EPA-600/8-84-020A Unsworth, M.H. 1982. Exposure to gaseous pollutants and uptake by plants. In E f f e c t s of Gaseous A i r P o l l u t i o n i n Ag r i c u l t u r e and H o r t i c u l t u r e (edited by Unsworth M.H. and Ormrod D.P.) pp43-63. Butterworth S c i e n t i f i c , London. Wilson, R.G., J.B. M i l l s and E.P. Wituschek. 1984. a report on the assessment of photochemical oxidants i n the Lower Mainland. Queen's Print e r for B.C., V i c t o r i a . 196 Appendix 1. Fresh weight - dry weight conversion factors for a l l species grown in 1985. Species Fresh weight / Dry weight conversion B r o c c o l i 8.42 X Bean 9.87 X Pea 5.62 X Carrot 9.13 X Dry weight = Fresh weight (g) Dry weight = Fresh weight (g) Dry weight = Fresh weight (g) Dry weight = Fresh weight (g) l 198 Appendix 2. Summary of exposure s tat is t ics for each species for the 1985 growing season. Summary of 1985 exposure s t a t i s t i c s . Species Cult T r t M7 M12 M14 Ml[7] Ml[12] Ml[14] AA BROC EMPEROR AA*1.5 SGI AA*2.0 AA*2.5 26.47 27.74 26.95 31.92 36.81 36.76 40.58 48.44 48.17 44.07 54.14 53.49 33.75 39.99 40.96 42.02 63.80 66.46 55.53 96.71 92.83 59.53 112.25 117.49 AA BEAN BBL-GV2 AA*1.5 GALAMORE AA*2.0 AA*2.5 AA PEA PUGET AA*1.5 BOLERO AA*2.0 AA*2.5 AA CARR HIPAK AA*1.5 SIXPAK AA*2.0 AA*2.5 26.09 26.95 25.88 34.13 38.34 37.33 43.56 50.20 48.64 47.67 56.87 55.10 26.14 26.94 25.83 34.22 38.33 37.27 43.66 50.10 48.48 47.80 56.85 55.01 25.45 26.06 24.91 34.06 37K.93 36.87 43.94 50.27 48.75 48.70 57.98 56.41 33.40 38.29 37.65 44.60 66.57 71.99 58.67 99.88 101.99 65.98 116.00 132.18 33.44 38.23 37.65 44.76 66.31 71.62 59.10 99.52 101.59 66.27 115.64 132.18 33.11 37.61 38.83 45.18 65.52 70.34 59.40 101.18 101.85 67.43 118.93 124.24 BROC = BROCCOLI; CARR = CARROT. Summary of 1985 exposure s t a t i s t i c s Species Cult T r t P7 P12 P14 PI[7] PI[12] PI[14] AA BROC EMPEROR AA*1.5 SGI AA*2.0 AA*2.5 AA BEAN BBL-GV2 AA*1.5 GAL AA*2.0 AA*2.5 AA PEA PUGET AA*1.5 BOLERO AA*2.0 AA*2.5 AA CARR HIPAK AA*1.5 SIXPAK AA*2.0 AA*2.5 47.89 55.66 47.44 62.93 95.92 85.23 91.47 132.80 115.87 79.87 137.83 117.28 47.89 55.66 48.95 62.93 95.92 85.23 91.47 132.80 115.87 94.65 137.83 117.28 47.89 55.66 48.95 62.93 95.92 85.23 91.47 132.80 115.87 94.65 137.83 117.28 47.89 55.66 48.95 62.93 95.92 85.23 91.47 132.80 115.87 94.65 137.83 117.28 72.58 121.66 121.66 79.70 199.15 199.00 139.38 433.13 433.13 98.08 429.20 429.20 72.58 121.66 121.66 80.51 292.98 292.98 139.38 433.13 433.13 129.40 429.20 429.20 72.58 121.66 121.66 80.51 292.98 292.98 139.38 433.13 433.13 129.40 429.20 429.20 72.58 121.66 121.66 95.03 292.98 292.98 139.38 433.13 433.13 129.40 429.20 429.20 BROC = BROCCOLI; CARR = CARROT; GAL = GALAMORE. 201 Appendix 3. Summary of exposure s t a t i s t i c s for each species for the 1986 growing season. Summary of 1986 exposure statistics for pea (153-211) Treatment M7 M12 Ml[7] Ml[l2] P7 P12 Pl[7] P l [ l 2 ] ppb ppb ppb ppb ppb ppb ppb ppb AA1 11 .59 13 .12 15 .97 18 .40 22 .64 28.03 45 .07 45 .88 AA2 15 .34 16 .90 21 .50 23 .83 37 .69 38.66 51 .84 84 .03 1B1 19 .40 22 .13 26 .88 34 .29 38 .60 43.34 66 .89 92 .82 2B1 22 .47 25 .53 31 .89 40 .36 59 .69 58.77 81 .47 132 .84 3B1 19 .08 21 .64 25 .83 33 .96 42.78 42.56 58 .49 105 .28 4B1 26 .30 30 .14 38 .05 48.51 56 .24 61.67 85 .20 112 .24 1B2 18 .42 21 .67 28 .14 41 .81 65 .49 64.49 90 .53 216 .48 2B2 21 .22 25 .32 34 .20 47 .56 47 .02 54.14 103 .73 131 .16 3B2 29 .18 33 .59 46 .14 59 .73 67 .47 79.15 113 .47 281 .22 4B2 33 .56 38 .96 51 .72 76 .01 74 .80 76.70 124 .49 411 .29 1B3 22 .82 24 .42 34 .06 42 .69 37 .29 47.68 152 .29 152 .29 2B3 30 .53 35 .51 44 .95 63 .64 69 .54 76.36 128 .12 258 .70 3B3 36 .81 41 .91 53 .58 72 .06 71 .88 88.08 156 .77 250 .08 4B3 34 .43 38 .37 48 .01 64 .53 58 .21 66.38 127.31 235 .94 Summary of 1986 exposure statistics for pea (153-211) Treatment SUM50 SUM80 SUM100 SUMALL IE50 IE80 IE100 ppb.h ppb.h ppb.h ppm.h ppb.h ppb.h ppb.h AA1 0 0 0 14 .53 0 0 0 AA2 12 .01 0 0 17.63 162 .01 0 0 1B1 145 .61 224 .90 164 .90 23 .98 1045 .61 464. 90 464 .90 2B1 538 .10 270.92 158 .60 25 .40 3038 .10 750. 92 558 .60 3B1 153 .22 83 .65 14 .68 22 .36 953 .22 483. 65 214 .68 4B1 144 .62 178 .70 121 .63 29 .43 5744 .62 738. 74 321 .63 1B2 1036 .96 924 .90 724.26 21 .34 3286 .96 1804. 89 1624 .26 2B2 971 .92 1485 .90 1203 .73 26 .45 3921 .92 3085. 88 2303 .73 3B2 3151 .32 1239 .20 682 .86 31 .22 10801 .32 4599. 22 2382 .86 4B2 3761 .93 1801 .82 1134 .90 37 .29 12511 .93 5721. 82 3534 .86 1B3 378 .69 250.98 130.73 26 .38 1778 .69 810. 98 530 .73 2B3 3054 .65 993 .83 528 .44 33 .70 10204 .65 4113. 83 1828 .44 3B3 4835 .35 1260 .04 1053 .58 37 .71 16135.35 5180.04 4053 .58 4B3 2424 .45 1350 .66 912 .90 38 .11 10424 .45 4150. 66 2412 .90 Summary of 1986 exposure statistics for pea (153-211) Treatment HGT80 HGT100 2C25 3C25 4C25 h h h h h AA1 0 0 7 5 3 AA2 0 0 15 14 11 1B1 3 3 39 32 27 2B1 6 4 38 35 32 3B1 5 2 28 22 18 4B1 7 2 46 43 41 1B2 11 9- 30 28 19 2B2 20 11 37 37 29 3B2 42 17 40 39 35 4B2 49 24 48 46 42 1B3 7 4 44 37 32 2B3 39 13 44 40 37 3B3 49 30 49 47 43 4B3 35 15 50 47 44 205 Summary of 1986 exposure s t a t i s t i c s f o r pea (153-211) Treatment D25 1 D50 D80 D25 2 D25 3 D25 4 days days days days days days AA1 16 0 0 4 1 1 AA2 27 2 0 13 1 1 1B1 47 9 0 23 13 3 2B1 45 19 1 22 13 3 3B1 40 10 1 19 6 3 4B1 51 28 6 26 16 4 1B2 37 15 4 16 12 2 2B2 46 22 8 22 10 4 3B2 47 33 16 21 14 5 4B2 51 40 21 24 20 4 1B3 50 14 2 25 15 4 2B3 51 34 16 25 15 4 3B3 51 39 20 25 19 5 4B3 53 41 16 26 20 4 Summary of 1986 exposure s t a t i s t i c s for pea (153-211) Treatment H10 H12 H14 GH10 GH12 GH14 GM12 ppb ppb ppb ppb ppb ppb ppb AA1 9.210 11 .17 14 .26 8. 155 9.694 12 .388 10.97 AA2 11 .82 16 .00 18 .45 9. 933 13.753 14 .856 13.71 1B1 15 .87 19.65 22 .11 13. 459 17.041 18 .557 18.01 2B1 17 .33 23 .72 26 .68 13. 434 19.046 20 .564 19.20 3B1 15 .45 19 .92 22 .18 12. 978 16.765 17 .490 17.33 4B1 20 .07 26 .62 31 .98 16.615 20.474 25 .206 23.20 1B2 14 .53 20 .12 23 .22 10. 867 14.246 17 .490 15.26 2B2 17 .68 21 .92 24 .91 13. 967 17.490 18 .737 18.70 3B2 22 .78 29 .97 35.91 16.036 21.807 25 .136 22.46 4B2 27 .81 34 .96 38 .31 21. 597 28.708 28 .543 28.60 1B3 19 .58 22 .84 25.74 16. 285 19.843 20 .464 20.32 2B3 23 .80 31.10 37 .10 18. 197 23.453 28 .347 25.41 3B3 29 .69 35 .57 31 .44 23. 025 26.564 33 .760 30.23 4B3 31 .44 34 .09 38 .62 25. 026 27.772 30 .726 29.56 Summary of exposure s t a t i s t i c s f o r potato (153-228,229) Treatment M7 M12 Ml[7] Ml[12] P7 P12 Pl[7] P l [ l 2 ] ppb ppb ppb ppb ppb ppb ppb ppb AA1 11 .62 13 .27 16 .11 17.87 23 .83 28 .11 46 .16 46 .16 AA2 14 .83 16 .51 20 .90 23 .48 37 .69 38 .66 51 .84 84 .03 1B1 19 .26 22.06 26 .50 33 .73 38 .60 43 .34 66 .89 92 .82 2B1 21 .98 24 .92 31 .53 39 .03 56 .69 58 .77 82 .82 132 .84 3B1 19 .59 21 .95 26 .69 34 .63 42.78 42 .56 58 .49 105 .28 4B1 27 .47 31 .26 38 .99 49 .24 56.24 61 .67 88 .31 115 .30 1B2 16 .97 20 .05 26 .45 38 .41 65 .49 64 .49 90 .53 216 .48 2B2 21 .62 25 .31 34 .05 46 .98 47 .02 54 .14 103 .73 131 .16 3B2 30 .10 33 .99 46.66 58 .55 73 .00 87 .24 125 .33 281 .22 4B2 32 .40 37.31 49 .00 71 .08 74 .80 76 .70 124 .49 411 .29 1B3 23 .57 26 .06 34.24 42 .36 37 .29 47 .68 152 .29 152 .29 2B3 31 .50 35 .59 45 .75 60 .66 69 .54 76 .36 136 .42 258 .70 3B3 37 .01 41 .21 54 .11 69 .22 74 .94 88 .08 156 .77 250 .08 4B3 34 .72 38 .43 47 .20 62 .81 62 .62 67 .39 134 .13 235 .94 208 Summary of exposure s t a t i s t i c s for potato (153-228,229) Treatment SUM50 SUM80 SUM100 SUMALL IE50 IE80 IE100 ppb.h ppb.h ppb.h ppm.h ppb.h ppb.h ppb.h AA1 0 0 0 19 .10 0 0 0 AA2 12 .01 0 0 22 .64 162 .06 0 0 1B1 147 .00 224.90 164 .90 30 .35 1097.00 464 .90 464 .90 2B1 609 .62 450.92 158 .57 31 .64 3359 .62 750 .92 558 .57 3B1 195 .87 83.65 14 .68 28 .58 1245 .87 483 .65 214.68 4B1 1443 .92 319.40 136 .90 38 .86 6993 .92 1119 .40 436 .93 1B2 1129 .72 924.89 724 .26 25 .61 3679 .72 1804 .89 1624 .26 2B2 1099 .96 1517.60 1215 .45 33 .35 4649 .96 3197 .60 2415 .45 3B2 3910.28 1571.60 793 .97 40 .16 13260 .20 5571 .57 3093 .97 4B2 4084 .30 1834.09 1134 .86 45.65 13934 .30 6154 .09 3534 .86 1B3 431 .31 250.98 130 .73 34 .64 2181 .31 810 .98 530 .73 2B3 3514 .75 1161.70 553 .19 43 .54 12114 .70 4681 .70 2053 .19 3B3 5674 .72 2046.60 1121.60 48 .34 19024 .70 6526 .57 4521 .60 4B3 2878 .42 1529.49 929 .68 48 .97 12428 .40 4729 .49 2629 .68 Summary of exposure statistics for potato (153-228,229) Treatment HGT80 HGT100 2C25 3C25 4C25 h h h h h AA1 0 0 9 7 5 AA2 0 0 17 15 12 1B1 3 3 49 40 33 2B1 6 4 48 42 36 3B1 5 2 36 28 21 4B1 10 3 62 58 56 1B2 11 9 33 30 20 2B2 21 12 47 44 33 3B2 50 23 55 51 47 4B2 54 24 61 57 52 1B3 7 4 58 50 44 2B3 44 15 60 54 50 3B3 56 34 66 63 58 4B3 40 17 68 64 61 Summary of exposure s t a t i s t i c s f o r potato (153-228,229) Treatment D25 1 D50 D80 D25 2 D25 3 D25 4 (153-228/9) (153-177) (178-211) (212-229) days days days days days days AA1 22 0 0 3 3 2 AA2 31 2 0 15 15 3 1B1 62 10 0 21 16 10 2B1 60 20 1 20 20 15 3B1 56 13 1 22 26 13 4B1 67 33 8 20 18 10 1B2 45 17 4 17 11 8 2B2 62 26 9 11 4 2 3B2 63 40 19 23 21 14 4B2 67 50 23 23 21 16 1B3 68 18 2 23 26 17 2B3 69 41 17 24 26 18 3B3 69 48 23 21 18 10 4B3 71 52 19 24 22 16 211 Summary of exposure s t a t i s t i c s f o r potato (153-228,229) Treatment H10 H12 H14 GH10 GH12 GH14 GM12 ppb ppb ppb ppb ppb ppb ppb AA1 9.25 11 .00 14 .60 8 .171 9.658 12.823 11 .24 AA2 11 .31 15 .12 18 .15 9 .710 13 .204 15.097 13 .71 1B1 16 .00 19 .24 22 .23 13 .964 17 .049 19.235 18 .51 2B1 16 .82 22 .22 26 .65 13 .747 18.437 21,262 19 .57 3B1 15 .85 20 .03 22 .92 13 .715 17 .502 18.863 18 .18 4B1 21.66 27 .58 33 .20 18 .523 22 .151 27.202 25 .03 1B2 13 .11 18 .03 21 .98 9 .858 12 .759 16.482 14 .17 2B2 18 .22 22 .06 25 .40 14 .788 17 .828 19.665 19 .33 3B2 23 .81 30 .64 36 .76 17 .531 23 .030 26.990 23 .98 4B2 26 .63 33 .35 37 .31 21 .335 27 .265 28.860 28 .41 1B3 20 .02 23 .47 27 .08 17 .136 20 .797 22.387 21 .55 2B3 24 .53 31 .46 38 .77 19 .521 24 .694 30.318 26 .90 3B3 29 .39 35.70 45 .23 23 .785 27 .823 35.416 31 .29 4B3 30 .93 33 .79 39 .30 25 .621 28 .635 32.322 30 .92 212 Appendix 4. Summary of frequency d i s t r i b u t i o n r e s u l t s f o r the ambient a i r 1 data, 1986 using MAXFIT. 213 Table A4-1. Descriptive statistics for Ambient Air 1 (2 minute average, 14-h, ppm) ozone data, 1986. Minimum 0.0014 Mean 0 .0124 Index of 1 .1311 skewness Maximum 0.0462 Variance 0 .0000 Kurtosis 5 .2589 Number of 1762 Standard 0 .0070 Skewness 1 .2794 observations deviation squared Table A4-2. Distribution parameters of Ambient Air 1 data (ppm). Distributions Normal Lognormal Gamma Weibull Johnsons S B Parameters Mean Variance Standard deviation Alpha Beta C Lower bound Upper bound 0.0124 0.0000 0.0070 -4.0482 0.1348 0.3672 -0.0062 -1.8058 0.2963 3.2344 0.0040 0.0126 1.6457 -0.0004 0.0011 -0.0030 0.0976 Table A4-3. Summary of the distribution results using Ambient Air 1 data (14-h, 2 min ave.) from 1986. Distributions Normal Lognormal Gamma Weibull Johnsons S B Significance Tests Absolute deviation 355.77 (5) 218.97 (1) 272.73 (3) 286.20 (4) 260.91 (2) Weighted abs. deviation 51.94 (3) 39.00 (1) 52.71 (4 55.46 (5) 48.69 (2) Chi-square significance .0 (5) .0000009 .0 (1) (4) .0 (5) .0 (2) Kolmogorov -Smirnov 94.97 (5) 37.99 (1) 57.68 (3) 63.81 (4) 49.80 (2) Cramer-Von Mises-Smirnov 2.20 (5) 0.49 (1) 0.85 (3) 0.96 (4) 0.69 (2) Log Likelihood 6236.20 (5) 6397.95 (4) 6405.99 (2) 6415.94 (1) 6400.18 (3) NB. Numbers in brackets are the rankings for each test performed. 215 Table A4-4. Descriptive statistics for Ambient Air 1 (2 minute average, 24-h, ppm) ozone data, 1986. Minimum 0.0014 Mean 0 .0105 Index of 1.3057 skewness Maximum 0.0521 Variance 0 .0000 Kurtosis 5.8127 Number of 3074 Standard 0 .0068 Skewness 1.7049 observations deviation squared Table A4-5. Distribution paramters of Ambient Air 1 data (ppm). Distributions Normal Lognormal Gamma Weibull Johnsons S B Parameters Mean Variance Standard deviation Alpha Beta C Lower bound 0.0105 -4.6283 0.0000 0.3519 0.0068 0.5932 -1.7986 0.8209 -0.0011 1.6771 0.0055 0.0099 1.3520 0.0012 0.0013 0.0005 Upper bound 0.0577 Table A4-6. Summary of the distribution results using Ambient Air 1 data (24-h, 2 min ave.) from 1986. Distributions Normal Lognormal Gamma Weibull Johnsons S B Significance Tests Absolute 748.39 611.38 460.68 355.22 528.02 deviation (5) (4) (2) (1) (3) Weighted abs. 130.87 161.63 117.93 88.66 131.40 deviation (3) (5) (2 (1) (4) Chi-square .0 .0 .0 .00000002 .0 significance (5) (4) (2) (1) (3) Kolmogorov 203.22 140.91 119.15 77.63 126.67 -Smirnov (5) (4) (2) (1) (3) Cramer-Von 5.03 3.67 2.19 1.01 2.56 Mises-Smirnov (5) (4) (2) (1) (3) Log 10993.76 11470.58 11542.56 11558.61 11518.00 Likelihood (5) (4) (2) (1) (3) NB. Numbers in brackets are the rankings for each test performed. 217 Table A4-7. Descriptive statistics for Ambient Air 1 (1 hour average, 24-h, ppm) ozone data, 1986. Minimum 0.00002 Mean 0 .0104 Index of 1 .2179 skewness Maximum 0.04243 Variance 0 .0000 Kurtosis 5 .2647 Number of 1749 Standard 0 .0066 Skewness 1 .4832 observations deviation squared Table A4-8. Distribution paramters of Ambient Air 1 data (ppm). Distributions Normal Lognormal Gamma Weibull Johnsons S B Parameters Mean Variance Standard deviation Alpha Beta C Lower bound 0.0104 0.0000 0.0066 -4.5020 0.2695 0.5191 -0.0022 -1.7743 0.5207 2.6615 0.0040 0.0118 1.6664 -0.0003 -0.0000 -0.0009 Upper bound 0.0674 Table A4-9 Summary of the distribution results using Ambient Air 1 data (24-h, 1-h ave.) from 1986. Distributions Normal Lognormal Gamma Weibull Johnsons S B Significance Tests Absolute 529.89 205.16 186.45 277.65 169.37 deviation (5) (3) (2) (4) (1) Weighted abs. 80.86 39.17 38.76 41.42 30.80 deviation (5) (3) (2) (4) (1) Chi-square .0 .0000036 .000033 .0 .00002 significance (5) (3) (1) (4) (2) Kolmogorov 128.97 65.22 35.25 54.60 43.60 -Smirnov (5) (4) (1) (3) (2) Cramer-Von 3.837 0.780 0.301 0.302 0.354 Mises-Smirnov (5) (4) (1) (2) (3) Log 6297.33 6539.00 6547.40 6528.75 6550.61 Likelihood (5) (3) (2) (4) (1) NB. Numbers in brackets are the rankings for each test performed. 

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.831.1-0097856/manifest

Comment

Related Items